(CODUX-107) Render invocation-bound planning plans

[numnx]

🧩 Task: Render invocation-bound planning plans

Automated task PR opened by Code UX for T02.

📋 Summary

Sprint: Sprint 175 · Fix Execution Plan · Task: T02

Ensure each Chat invocation for sprint planning displays the execution plan persisted for that invocation's own sprint, not a generic or most-recent planning entry.

🤖 Model & Provider

Provider	Codex CLI
Model	gpt-5.5

⏱️ Timing

Started	2026-07-07 12:01:08 UTC
Finished	2026-07-07 12:09:33 UTC
Duration	8m 24s

📊 Token Usage

Metric	Value
Input tokens	121,571
Cached input tokens	3,465,856
Output tokens	16,909
Total tokens	3,604,336
Invocations	1
Tool calls	68
Active time	7m 38s

Included usage estimate (subscription/local login): $2.85

✅ QA Review

⏳ QA review runs after this PR is opened. Check the task card in the dashboard for the outcome once it lands.

📝 Full Task Prompt

## Objective
Render sprint-specific execution plan details in Chat invocation transcripts using the metadata persisted on the selected planning invocation message.

## Scope
- Edit `dashboard/src/v2/lib/chat-widget-view-models.ts`
- Edit `dashboard/src/v2/components/chat/widgets/PlanningRequestWidget.tsx`
- Edit `dashboard/src/v2/components/chat/InvocationMessageBubble.tsx` only if prop plumbing is needed
- Update `tests/dashboard/lib/chat-widget-view-models.test.ts`
- Update `tests/dashboard/v2/chat-message-bubbles.test.tsx`
- Relevant symbols: `ChatWidgetState`, `getInvocationWidgetData`, `PlanningRequestWidget`, `InvocationMessageBubble`

## Implementation Requirements
1. Extend the widget view model to safely parse optional `metadata.executionPlan` fields: `sprintId`, `sprintNumber`, `sprintName`, `goal`, `taskCount`, `createdTaskIds`, and task summaries.
2. Prefer execution-plan metadata over the current fallback `Execution Plan` label when rendering planning invocation messages, while keeping the existing fallback for legacy messages without metadata.
3. Render the planning widget with the sprint name/key and a compact task summary so two planning invocations for different sprints are visually distinguishable in the transcript.
4. Keep the widget as an adjunct to the markdown bubble; do not replace or hide the underlying message content.
5. Add tests showing that two invocation messages with different `executionPlan` metadata render different sprint names/task summaries, and that legacy virtual-route metadata still renders the existing fallback without throwing.

## Constraints
- Do not fetch current sprint state from the dashboard to populate this widget; it must use the selected invocation message metadata.
- Do not change the invocation list selection or message cache keys unless a focused failing test proves it is necessary.
- Preserve accessibility basics: status text must remain readable to screen readers and interactive elements must not be nested inside the message bubble unexpectedly.
- Keep existing chat widget tests for generic planning and virtual-route messages passing.

## Verification
- Run `pnpm vitest run tests/dashboard/lib/chat-widget-view-models.test.ts tests/dashboard/v2/chat-message-bubbles.test.tsx`.
- Success means the planning widget displays sprint-specific execution plans from message metadata and legacy planning widgets still render safely.

🌿 Branch Info

Base: feature/CODUX-175-fix-execution-plan
Head: task/feature-codux-175-fix-execution-plan-t02-codex-mrallibr

---

🤖 Generated by Code UX