docs(features): catch up the features page with several releases of shipped work (#392)

* docs(features): catch up the features page with several releases of shipped work

The features.md page on docs.testplanit.com had fallen 80+ feature
and enhancement merges behind. Refreshes every section against the
current product surface and adds three new ones:

- Parameterized Test Cases (parameters, local + shared datasets,
  iterations, sensitive values, exclude-draft-from-runs toggle)
- Review and Approval (gated workflow states, reviewer assignment,
  strict transitive gates, reminders, system + per-project toggles)
- Result Governance (required fields, edit window, flip
  justification, linked-issue-on-failure)
- Webhooks and Extensibility as a top-level section covering
  outbound events, inbound issue-tracker sync, and the MCP server
  for agent integrations

Existing sections gained entries for items shipped since the prior
revision: immutable completed runs, SSE-driven live updates, PDF
exports, GitLab + Gitea + GitHub Enterprise integrations, OAuth 2.0
for issue trackers, generate-from-URL, Mobilewright templates,
prompt configurations, additional locales, self-registration toggle,
expanded audit coverage, API tokens, multi-tenant + external-DB
deployment options, configurable JUnit iteration-property mapping,
folder + tag report dimensions, Execution Log + Automation
Candidates + Parameter Iteration Matrix reports, cross-project
reports, and more.

No behavior changes; documentation only.

* docs(features): add Markdown Parsing, Duplicate Detection, Automation Candidates LLM ranking

Caught three shipped LLM features that the prior commit missed:
markdown-parsing assist on the import surface, semantic duplicate
detection layered on the syntactic scan, and the LLM ranking that
drives the Automation Candidates report. Also names the generic
OpenAI-compatible custom-endpoint provider explicitly so the
provider list is no longer using "and more" for a shipped option.

Author: therealbrad

docs/docs/features.md

@@ -23,18 +23,52 @@ TestPlanIt is a comprehensive test management platform designed to help teams pl
 - **Custom fields** - Filter test cases by custom fields
 - **Tags** - Apply tags to categorize and filter test cases across projects
 - **Issues** - Attach issues to quickly navigate between test cases and related issues
+- **External link attachments** - First-class link attachments alongside file uploads (paste a URL onto any drop zone)
 - **Version history** - Track changes to test cases over time with full version control
 - **Documentation pages** - Create wiki-style documentation pages within projects
+- **Duplicate detection** - Background scan surfaces near-duplicate test cases at create time and from a dedicated review surface
+- **Copy / move** - Reorganize cases across folders and projects, including in-place reordering
+
+### Parameterized Test Cases
+
+- **Named typed parameters** - Declare STRING / INTEGER / BOOLEAN / SELECT parameters on a case and reference them in steps via `@chip` mentions
+- **Local + shared datasets** - Each case can have its own owner-bound dataset, or assign a project-scoped shared dataset and map its columns to the case's parameter names
+- **Version pinning** - Pin a case to a specific shared-dataset version, or follow the latest
+- **Iterations** - Parameterized cases run once per dataset row in a test run, each with its own per-row status; the case-level status rolls up as worst-of
+- **CSV import** - Paste-CSV and full upload wizard for bulk dataset row authoring
+- **Sensitive values** - Mark a parameter sensitive to mask it in the UI and redact it in exports and notifications for users without read-sensitive permission
+- **Parameter Iteration Matrix report** - Pre-built report that pivots iteration outcomes across runs
+- **Exclude draft cases from runs** - Opt-in per-project setting that keeps Not Started cases out of the Add Cases picker and auto-removes them from open runs when reverted
+
+### Review & Approval
+
+- **Gated workflow states** - Mark any workflow state as "requires review"; transitions into it (or across it) need an approved review request
+- **Reviewer assignment** - Request review from a specific user or any holder of a role; reviewers see their queue in a Review Inbox
+- **Strict transitive gates** - A single transition crossing multiple gated states requires an approval for each one
+- **Decision audit** - All decisions, comments, and reminders are recorded and emitted as outbound webhook events
+- **Reminders** - Configurable hourly scheduler nudges reviewers when requests sit pending past a threshold
+- **System + per-project toggles** - System admins flip the master switch; project admins opt their project in independently
 
 ## Test Execution
 
 ### Test Runs
 
 - **Flexible run creation** - Create test runs from entire folders, filtered sets, or individual test cases
 - **Magic Select** - Create test runs from similar test cases using AI
-- **Configuration support** - Execute tests against different configurations (browsers, environments, etc.)
+- **Multi-configuration runs** - Execute tests against different configurations (browsers, environments, OS, etc.) in parallel; configurations are scoped to projects
 - **Bulk status updates** - Quickly update multiple test results at once
 - **Execution history** - View the complete history of test executions for any run or individual test case
+- **Immutable completed runs** - Once a run is marked complete its case set, results, and configuration are structurally frozen; admins can edit a single result via the audit-log-tracked override path
+- **Live updates** - SSE-driven push so multiple testers watching the same run see each other's status changes without refreshing
+- **PDF export** - Per-run PDF with expanded step details and per-iteration values; also available at the milestone level
+- **Forecasting** - Live estimates of remaining duration based on historical pass/fail per case
+
+### Result Governance
+
+- **Required result fields** - Optional per-project requirement that result-recording forms must collect certain fields
+- **Edit window** - Lock results from in-place edit after a configurable window; later edits go through the audited override path
+- **Flip justification** - Require a justification note when a result flips a previously-passed iteration to failed (or vice versa)
+- **Linked-issue requirement** - Optional per-project requirement that failures must be linked to an issue before save
 
 ### Automation Integration
 
@@ -43,6 +77,7 @@ TestPlanIt is a comprehensive test management platform designed to help teams pl
 - **TestPlanIt CLI** - CLI tool for submitting test case results in popular formats
 - **API access** - Push results programmatically via the REST API
 - **WebdriverIO reporter** - Native integration with WebdriverIO test framework
+- **JUnit iteration property mapping** - Configurable per-project list of JUnit property/attribute names the importer uses to detect the iteration index — defaults to `iteration`, fully customizable for any CI vocabulary
 
 ## Exploratory Testing
 
@@ -66,8 +101,13 @@ TestPlanIt is a comprehensive test management platform designed to help teams pl
 ### Issue Tracking Integration
 
 - **Jira integration** - Link test cases and results to Jira issues
-- **GitHub integration** - Connect with GitHub Issues for defect tracking
+- **GitHub integration** - Connect with GitHub Issues for defect tracking; supports GitHub Enterprise Server via configurable API base URL
+- **GitLab integration** - Link to GitLab.com or self-hosted GitLab issues
+- **Gitea / Forgejo / Gogs integration** - Connect to self-hosted Gitea-family issue trackers
 - **Azure DevOps integration** - Sync with Azure Boards work items
+- **OAuth 2.0 authentication** - Use OAuth for GitHub, GitLab, and Gitea/Forgejo connections (no personal access token required)
+- **Two-way sync** - Optional outbound notifications and inbound webhook updates between TestPlanIt and the issue tracker
+- **Linked-issue context for AI** - Auto-tag and test case generation pull labels, components, and metadata from linked issues
 - **Bi-directional linking** - Navigate seamlessly between tests and issues
 
 ## Reporting & Analytics
@@ -80,8 +120,12 @@ TestPlanIt is a comprehensive test management platform designed to help teams pl
 
 ### Reports
 
-- **Report builder** - Create custom reports with drag-and-drop interface
+- **Report builder** - Create custom reports with drag-and-drop interface, including folder and tag dimensions
 - **Test run reports** - Detailed breakdowns of test execution results
+- **Execution Log report** - Pre-built timeline of every result event with expanded step details
+- **Automation Candidates report** - LLM-ranked list of manual cases best suited for automation, drawing on custom field signals and attached-issue metadata
+- **Parameter Iteration Matrix** - Pivots per-iteration outcomes across runs for parameterized cases
+- **Cross-project reports** - Aggregate results across multiple projects in a single view
 - **Coverage analysis** - Understand what has been tested and what remains
 - **Trend analysis** - Track quality metrics over time
 - **Export options** - Export reports in CSV for stakeholders
@@ -95,57 +139,94 @@ TestPlanIt is a comprehensive test management platform designed to help teams pl
 
 ### LLM Integration
 
-- **Test case generation** - Generate test cases from requirements using AI
-- **QuickScript AI generation** - Convert manual test cases into automation scripts with AI, optionally informed by your code repository (GitHub, GitLab, Bitbucket, Azure DevOps, Gitea/Forgejo/Gogs)
+- **Test case generation** - Generate test cases from requirements using AI; optional `includeParameters` toggle (admin-gated) for parameterized output
+- **Generate from URL** - Crawl a webpage or sitemap and generate test cases targeting it
+- **Markdown Parsing** - Pasted Markdown is parsed by an LLM into structured cases (name, steps, expected results, custom field values) on the Import Markdown surface
+- **QuickScript AI generation** - Convert manual test cases into automation scripts with AI, optionally informed by your code repository (GitHub, GitLab, Bitbucket, Azure DevOps, Gitea/Forgejo/Gogs); ships with templates spanning web, mobile, API, and Mobilewright targets
 - **Enhance Writing** - Get AI recommendations to improve writing for any rich text field
 - **Magic Select** - AI-assisted test case selection for quickly building test runs
-- **Auto Tag** - Automatically suggest and apply tags to test cases, test runs, and sessions using AI analysis
-- **Multiple providers** - Support for OpenAI, Azure OpenAI, Anthropic, Ollama, and more
+- **Auto Tag** - Automatically suggest and apply tags to test cases, test runs, and sessions using AI analysis; incorporates linked-issue context (Jira labels + components, etc.) for sharper suggestions
+- **Duplicate Detection** - LLM-scored semantic similarity layered on top of the syntactic scan so near-duplicates with reworded steps are surfaced too
+- **Automation Candidates ranking** - The Automation Candidates report (under [Reports](#reports)) is LLM-ranked over manual cases + their custom fields + drilled-down issue metadata
+- **Prompt Configurations** - Per-feature prompt templates that admins can override system-wide or per project
+- **Multiple providers** - Support for OpenAI, Azure OpenAI, Anthropic, Google Gemini, Ollama, plus a generic OpenAI-compatible custom endpoint, with per-feature provider routing
 - **Privacy options** - Use local models for sensitive data with Ollama integration
+- **Capability probing + billing periods** - Admin UI surfaces each provider's available capabilities and tracks usage against custom billing periods
+
+### MCP Server
+
+- **Model Context Protocol** - First-class MCP server exposing TestPlanIt entities as resources and tools so agents (Claude Desktop, Cursor, etc.) can read and edit cases, sessions, runs, and results
+- **Auth-aware** - Operates within the calling user's permissions; honors all access policies the rest of the product enforces
 
 ## Administration
 
 ### User Management
 
-- **Role-based access** - Define custom roles with granular permissions
+- **Role-based access** - Define custom roles with granular permissions, including a per-area "Can Approve" right for Review & Approval reviewers
 - **Groups** - Organize users into groups for easier permission management
 - **SSO support** - Integrate with your identity provider via SAML, Google OAuth, Apple Sign In, or Microsoft (Azure AD)
+- **Magic Link sign-in** - Optional passwordless email link as an alternative to password authentication
+- **Self-registration toggle** - Admin-controlled switch for letting users create their own accounts (or restricting account creation to admins)
 
 ### Customization
 
 - **Custom workflows** - Define status workflows that match your process
 - **Custom statuses** - Create statuses with custom icons and colors
 - **Templates** - Create templates for consistent test case structure
-- **User preferences** - Configure theme, locale, timezone, and date/time formats
+- **User preferences** - Configure theme, locale, timezone, and date/time formats; preference selections are visible from the profile view as well as the editor
+- **Configurations** - Author OS / browser / environment configurations scoped to projects, with an admin UX overhaul covering bulk assign and search
 
 ### Security & Compliance
 
 - **Password policy** - Configure minimum length, character requirements, history depth, and expiration
 - **Account lockout** - Protect against brute-force attacks with configurable threshold and duration
 - **Password enforcement** - Force password changes (individual or bulk) and revoke passwords
 - **Password strength indicator** - Real-time zxcvbn-powered feedback on signup and password change forms
-- **Audit logs** - Track all changes for compliance and security review
+- **Audit logs** - Track all changes for compliance and security review; covers admin configuration mutations, system-actor attribution, and a tamper-evident archive surface for long-term retention
 - **Two-factor authentication** - Add an extra layer of security for user accounts
 - **Data encryption** - Secure data at rest and in transit
+- **API tokens** - Personal and service-account API tokens with scoped access and revocation
 
 ## Collaboration
 
 ### Team Features
 
 - **Comments** - Discuss test cases and results with your team
-- **Notifications** - Stay informed about changes and assignments
+- **Notifications** - Stay informed about changes and assignments via in-app inbox, email, or daily digest
+- **Real-time notification delivery** - Server-sent events push the bell badge instantly, no polling
 - **@mentions** - Tag team members in discussions
 - **Activity feeds** - See recent activity across your projects
+- **Localized UI** - Ship-with-product localization for 13 locales including English, German, Spanish, French, Italian, Japanese, Korean, Dutch, Polish, Portuguese (Brazil), Russian, Turkish, Vietnamese, and Simplified + Traditional Chinese
 
 ### Import & Export
 
 - **Drag and drop import** - Drag files directly from your desktop onto the Repository or Test Runs page to instantly start an import with the file pre-loaded
 - **Bulk import** - Import test cases from CSV with automatic markdown/HTML detection for rich text fields
+- **Markdown import** - Paste markdown source directly to seed cases, steps, and documentation pages
 - **Test Results import** - Import automated test results via popular formats like JUnit XML
 - **TestMo migration** - Special import support for migrating from Testmo
 - **Export capabilities** - Export data in CSV or PDF, with markdown format option for rich text fields
+- **Data Lake Export** - Project-scoped export endpoints that emit normalized JSON for warehouse pipelines
 - **API access** - Full programmatic access to all features
 
+## Webhooks & Extensibility
+
+### Outbound Webhooks
+
+- **Per-project endpoints** - Configure webhook destinations from project settings, with HMAC signing and optional Slack adapter
+- **Event catalog** - Subscribe to fine-grained events: case lifecycle, run completion, per-iteration results, session updates, review requests, review decisions, reviewer reminders, and more
+- **Per-iteration payloads** - `iteration.result.recorded` event ships per-row values with sensitive parameters redacted
+- **Replay + audit** - Failed deliveries are queued, retried with exponential backoff, and replayable from the admin surface; every delivery is audited
+
+### Inbound Webhooks
+
+- **Issue-tracker sync** - Configure inbound webhooks from Jira / GitHub / GitLab / Gitea so issue updates flow back into TestPlanIt
+- **Adapter-aware** - The inbound config is keyed to the issue integration so swapping providers cleanly resets the webhook
+
+### Agent Integrations
+
+- **MCP Server** - See [AI-Powered Features → MCP Server](#mcp-server) — first-class Model Context Protocol surface so Claude Desktop, Cursor, and similar agents can read and edit TestPlanIt entities through the same access policies as the UI
+
 ## Search & Discovery
 
 - **Advanced search** - Powerful search across all entities with complex filters
@@ -155,6 +236,10 @@ TestPlanIt is a comprehensive test management platform designed to help teams pl
 
 ## System & Operations
 
-- **Background job processing** - Asynchronous task processing with BullMQ
+- **Background job processing** - Asynchronous task processing with BullMQ (workers cover audit, search reindex, email, webhooks, AI, dataset import, and more)
 - **System health monitoring** - Monitor system status and queue health
+- **Multi-tenancy support** - Optional schema-isolated multi-tenant mode for SaaS deployments
+- **External database support** - Run against your own managed PostgreSQL, Redis-compatible cache (Redis/Valkey), and Elasticsearch instances
+- **Docker-first deployment** - Production Compose files, recommended ARM + x86 images, and a documented manual setup path
 - **Upgrade notifications** - Get notified about new features since the last time you logged in
+- **Release notes catalog** - In-app catalog of version-specific marketing copy with one-click jumps to the relevant documentation