XS⚠️ ◾ Fix wizard heading icon and align "What we collect" icons in telemetry consent dialog

[tomek-i]

Summary

The "Help Us Improve YakShaver" telemetry consent dialog (shown early in the setup flow) had inconsistent iconography:

• The heading used a generic BarChart3 "analyze" icon instead of the YakShaver brand icon.
• The three "What we collect" list icons (Error reports, Workflow performance, Usage metrics) used an ad-hoc mt-0.5 top offset that didn't reliably line up with their checkbox/label row, since row heights differ (the row is min-h-11, the icon offset was a fixed guess).

Changes

• src/ui/src/components/settings/telemetry/TelemetryConsentDialog.tsx
  • Replaced the <BarChart3 className="h-5 w-5" /> heading icon with the YakShaver (small Red Bull) logo (/logos/SQ-YakShaver-LogoIcon-Red.svg), the same asset already used elsewhere in the app (e.g. the onboarding sidebar and recording control bar).
  • Wrapped each "What we collect" icon (Bug, Activity, BarChart3) in a flex items-center min-h-11 container that matches the height of its corresponding checkbox/label row, replacing the fixed mt-0.5 offset. This centers each icon against its row instead of guessing a top margin.

Acceptance criteria mapping

• The icon next to the "Help Us Improve YakShaver" heading uses the YakShaver (small Red Bull) icon → heading now renders SQ-YakShaver-LogoIcon-Red.svg.
• The icons for the "What we collect" items are visually aligned with each other and with their corresponding text → each icon is now vertically centered against its checkbox/label row via a matching min-h-11 container, so all three align consistently.

Testing performed

• npm run build — passes (tsc clean).
• npm run lint — passes (0 errors; one pre-existing, unrelated info-level suggestion in Cloud360LiveView.tsx left untouched, out of scope).
• npm run format then git diff --exit-code — clean, no formatting drift.
• npm rebuild better-sqlite3 --build-from-source && npx vitest run --exclude 'src/ui/**' — 696/696 backend tests pass.
• npm --prefix src/ui test — 261/261 UI tests pass.

This is a scoped, presentational-only change (icon swap + alignment wrapper) with no behavioural/state changes, so no new tests were added.

Closes #964

[github-actions]

PR Metrics

✔ Thanks for keeping your pull request small.
⚠️ Consider adding additional tests.

	Lines
Product Code	11
Test Code	-
Subtotal	11
Ignored Code	-
Total	11

Metrics computed by PR Metrics. Add it to your Azure DevOps and GitHub PRs!

[Copilot]

Pull request overview

Updates the telemetry consent dialog UI to use YakShaver-branded iconography and improve visual alignment in the “What we collect” section.

Changes:

• Swapped the dialog title icon from BarChart3 to the YakShaver logo asset (SQ-YakShaver-LogoIcon-Red.svg).
• Replaced fixed mt-0.5 icon offsets with a flex items-center min-h-11 wrapper to vertically center each “What we collect” icon against its corresponding checkbox/label row.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

🚀 Pre-release build is available for this PR:
https://github.com/SSWConsulting/SSW.YakShaver.Desktop/releases/tag/0.6.0-beta.991.1784632239

[tomek-i]

🔭 crows-nest: ready-PR pipeline started — review → address → re-validate → gated merge.

[tomek-i]

🔭 muster review — PR #991

Verdict: no blocking findings — review advisory only.

Severity	Count
blocking	0
major	0
minor	0
nit	0

⚠️ Degraded review — only one lens ran (code-review). The codex-rescue second-opinion lens is not available as an agent type in this environment (no nested/plugin agent types beyond the generic set), so this is a single-lens read, not the full two-lens muster review.

The code-review lens checked: dangling BarChart3 import, consistency of the ?url logo-import pattern with other usages in the codebase, icon sizing/color parity with the lucide icons it replaces, correctness of the flex items-center min-h-11 alignment fix across all three rows, and the alt="" decorative-image choice. No issues found — the change is a small, scoped, presentational fix consistent with the PR's stated testing (build/lint/format clean, 696 backend + 261 UI tests passing).

Bottom line: no blocking findings — advisory pass only, given the single-lens degrade.

[tomek-i]

🔭 crows-nest: blocked — review degraded, no human sign-off on the automated verdict.

The merge gate (5-point, computed deterministically from the run state) returned blocked:

• review degraded (a lens failed) — a missing review is not a green light

Why it's degraded: only the code-review lens could run in this environment. The codex-rescue second-opinion lens (agentType: codex:codex-rescue) is not registered as an available agent type here, so muster's dual-lens review fell back to a single lens per its documented fallback (muster §1/§5). The one lens that did run found zero findings (see the review comment above) — the change looks clean — but per crows-nest's merge-gate design, a degraded review is treated as not safe to merge regardless of autoMerge, since a missing second opinion is never inferred as "no findings ⇒ safe."

Gate state fed to merge-gate.mjs: CI green, local checks assumed green (matches CI + the PR's own reported test run), not draft, MERGEABLE, 0 blocking findings, degraded review.

This needs a human. A maintainer should read the diff (small, self-contained icon/alignment fix, 1 file) and merge manually if satisfied — the automation can't clear this on its own until a second review lens is available in this environment.

Note for the operator: every other queued PR in this fleet run will hit this same structural block (missing codex:codex-rescue agent type), not a defect in these PRs themselves.