Test RAG retry telemetry logging

[BigSimmo]

Summary

• add answer retry count/reason telemetry to RAG answer timings and query-log metadata
• cache packed generation context during retry attempts using document-scope-aware keys
• add regression coverage for retry reasons and packed context key reuse

Verification

• npm run test -- tests/rag-answer-fallback.test.ts tests/rag-context-budget.test.ts

[Copilot]

Pull request overview

Adds retry telemetry and context-pack caching to the RAG answer generation path, and introduces regression tests to ensure retry reasons are recorded and context packing can be reused safely across retries.

Changes:

• Extend RagAnswer.latencyTimings to include context_pack_cache_hits, answer_retry_count, and answer_retry_reasons.
• Add a stable packedContextCacheKey() and a per-answer in-memory cache to reuse packed context across retry attempts.
• Add/extend tests to validate retry telemetry recording and packed-context cache key stability (including document-scope sensitivity).

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated no comments.

File	Description
tests/rag-context-budget.test.ts	Adds coverage for packedContextCacheKey() stability and document-scope behavior.
tests/rag-answer-fallback.test.ts	Adds regression test for retry telemetry logging and asserts retry timing fields in an existing fallback scenario.
src/lib/types.ts	Extends RagAnswer.latencyTimings with new retry/cache telemetry fields.
src/lib/rag.ts	Implements stable cache keys + context-pack reuse across retries; records retry count/reasons into answer timings and query-log metadata.

---

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

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 1515d239bc

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Preserve retry counters in fallback logs

When generation falls into this fallback path after a retry has already been scheduled (for example, the fast answer triggers fast_*_retry_strong and the strong model call throws), these counters are present on the returned fallbackAnswer.latencyTimings, but the logRagQuery metadata built in the catch block still only writes context_pack_latency_ms and omits context_pack_cache_hits, answer_retry_count, and answer_retry_reasons. That makes the production rag_queries telemetry lose the retry reason exactly for failed retry attempts, so the new retry logging is incomplete for generation-fallback cases.

Useful? React with 👍 / 👎.