fix(reports): split matrix reports to rows × columns (ADR-0021 D2) (#382)

The four matrix reports flattened their across-dimension into `rows` (the
pre-9.1 stopgap, which the dataset renderer degraded to a flat grouped
table). spec 9.1 added `Report.columns` for the true cross-tab axis, so
move the second dimension into `columns`:

  - accounts_by_industry_type:   industry × type
  - lead_inflow_by_month_source: lead_source × last_contacted_date
  - cases_opened_by_day_priority: priority × created_date
  - pipeline_coverage_by_quarter: forecast_category × close_date

The two summary reports keep their multi-level `rows` (legit down-axis
nesting, no across). Refresh three doc comments that referenced the
retired `groupingsDown`/`groupingsAcross` fields. Bumps @objectstack/spec
to ^9.2.0 (the `columns` field landed in 9.1).

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

Author: os-zhuang

package.json

@@ -36,7 +36,7 @@
     "@objectstack/runtime": "^9.0.0",
     "@objectstack/service-analytics": "^9.0.0",
     "@objectstack/service-automation": "^9.0.0",
-    "@objectstack/spec": "^9.0.0"
+    "@objectstack/spec": "^9.2.0"
   },
   "optionalDependencies": {
     "better-sqlite3": "^12.10.0"

pnpm-lock.yaml

@@ -6,7 +6,7 @@ export const AccountsByIndustryTypeReport: ReportInput = {
   name: 'accounts_by_industry_type',
   label: 'Accounts by Industry and Type',
   description: 'Matrix report showing accounts by industry and type',
-  dataset: 'account_metrics', rows: ['industry', 'type'], values: ['account_count', 'annual_revenue_sum'],
+  dataset: 'account_metrics', rows: ['industry'], columns: ['type'], values: ['account_count', 'annual_revenue_sum'],
   type: 'matrix',
   runtimeFilter: { is_active: true },
 };

src/reports/account.report.ts

@@ -31,6 +31,6 @@ export const CasesOpenedByDayPriorityReport: ReportInput = {
   name: 'cases_opened_by_day_priority',
   label: 'Cases Opened by Priority × Day',
   description: 'Daily case inflow split by priority',
-  dataset: 'case_metrics', rows: ['priority', 'created_date'], values: ['case_count'],
+  dataset: 'case_metrics', rows: ['priority'], columns: ['created_date'], values: ['case_count'],
   type: 'matrix',
 };

src/reports/case.report.ts

@@ -13,6 +13,6 @@ export const LeadInflowByMonthSourceReport: ReportInput = {
   name: 'lead_inflow_by_month_source',
   label: 'Lead Engagement by Month × Source',
   description: 'Contacted-lead volume per month, broken down by acquisition channel',
-  dataset: 'lead_metrics', rows: ['lead_source', 'last_contacted_date'], values: ['lead_count'],
+  dataset: 'lead_metrics', rows: ['lead_source'], columns: ['last_contacted_date'], values: ['lead_count'],
   type: 'matrix',
 };

src/reports/lead.report.ts

@@ -30,23 +30,23 @@ export const WonOpportunitiesByOwnerReport: ReportInput = {
 /**
  * Quarterly pipeline coverage — the matrix view that powers the classic
  * sales-ops "pipeline coverage" conversation: each forecast category against
- * the quarter the deal is expected to close in. Uses the spec's
- * `dateGranularity` on `groupingsAcross` so the server can bucket close_date
- * into quarters in a single aggregate query.
+ * the quarter the deal is expected to close in. `close_date` is the across
+ * dimension (`columns`); the dataset's `dateGranularity` buckets it into
+ * quarters in a single server-side aggregate query.
  */
 export const PipelineCoverageByQuarterReport: ReportInput = {
   name: 'pipeline_coverage_by_quarter',
   label: 'Pipeline Coverage by Forecast × Quarter',
   description: 'Open pipeline amount by forecast category, bucketed by close quarter',
-  dataset: 'opportunity_metrics', rows: ['forecast_category', 'close_date'], values: ['total_amount', 'opp_count'],
+  dataset: 'opportunity_metrics', rows: ['forecast_category'], columns: ['close_date'], values: ['total_amount', 'opp_count'],
   type: 'matrix',
   runtimeFilter: { stage: { $nin: ['closed_won', 'closed_lost'] } },
 };
 
 /**
  * Sales-rep funnel — two-level summary that mirrors how reps drill into their
- * own book of business. Tests the multi-level `groupingsDown` rendering and
- * the `aggregate`-mixed-with-`detail` column pattern.
+ * own book of business. Exercises multi-level `rows` grouping (owner → stage)
+ * over the dataset's measures.
  */
 export const OpportunityFunnelByOwnerStageReport: ReportInput = {
   name: 'opportunity_funnel_owner_stage',