feat(rag): make the RAG embed batch size configurable (#4327)

[JSONbored]

Summary

• EMBED_BATCH (items per RAG embed-provider call) was a hardcoded 96 in src/review/rag.ts with no env override, unlike its sibling RAG_DIMENSIONS/QDRANT_DIM. Adds AI_EMBED_BATCH, threaded through RagInfra the exact same way embeddingDimensions already is, so self-host operators can tune throughput for their own hardware without a code change.
• No default change. Benchmarked the actual embedding path on an RTX A5000 (real chunk text, matching the production batch size) as part of GPU passthrough for Ollama + embedding benchmark vs. the current CPU-bound setup #4327: 96 is already near-optimal (~34ms/chunk vs ~70ms/chunk at 32; only ~3% further gain going all the way up to 384). This PR is the tuning-knob half of that issue's recommendation, not a default change.

Part of #4327.

Scope

• The PR title follows type(scope): short summary Conventional Commit format.
• This PR is focused and does not mix unrelated backend, UI, MCP, docs, dependency, and deploy changes.
• This follows CONTRIBUTING.md and does not reintroduce GitHub Pages, VitePress, site/, or CNAME.
• Part of GPU passthrough for Ollama + embedding benchmark vs. the current CPU-bound setup #4327 (maintainer-owned infra/GPU-migration epic tracking issue; this PR does the code-side follow-up recommended there).

Validation

• git diff --check
• npm run actionlint (via npm run test:ci)
• npm run typecheck (via npm run test:ci)
• npm run test:coverage locally — rag.ts 100% branch coverage (174/174) including the new code; adapters.ts's one pre-existing partial branch (line 43) is unrelated code this PR does not touch
• npm run test:workers (via npm run test:ci)
• npm run build:mcp (via npm run test:ci)
• npm run test:mcp-pack (via npm run test:ci)
• npm run ui:openapi:check (via npm run test:ci) — no API/OpenAPI surface touched
• npm run ui:lint (via npm run test:ci) — no UI files touched
• npm run ui:typecheck (via npm run test:ci)
• npm run ui:build (via npm run test:ci)
• npm audit --audit-level=moderate — 0 vulnerabilities
• New behavior (the embedBatch override) has unit tests covering both the ragEmbedBatchFromEnv parser branches and the plumbing through embedTexts, upsertChunks, retrieveContext, and createReviewAdapters

Ran npm run test:ci twice end-to-end (full unsharded suite) to confirm green; one unrelated pre-existing flaky test (queue.test.ts's #3899 concurrency-bound regression, nothing to do with RAG/embeddings) failed on the first run under heavy background load and passed both in isolation and on a clean re-run.

Safety

• No secrets, wallet details, hotkeys, coldkeys, user PATs, private keys, raw trust scores, private rankings, or private maintainer evidence are exposed.
• Public GitHub text stays sanitized, low-noise, and does not imply compensation guarantees or optimization tactics.
• Auth, cookie, CORS, GitHub App, Cloudflare, or session changes include negative-path tests. — N/A, no such surface touched.
• API/OpenAPI/MCP behavior is updated and tested where needed. — N/A, no API/OpenAPI/MCP surface touched.
• UI changes use live API data or real empty/error/loading states. — N/A, backend-only change.
• Visible UI changes include a UI Evidence section. — N/A, no UI change.
• Public docs/changelogs are updated where needed (.env.example documents the new var); changelog itself not touched (not a release-prep PR).

Notes

• No migration, OpenAPI, or cf-typegen regeneration needed — AI_EMBED_BATCH is a plain env.d.ts field read directly in src/review/adapters.ts, outside the scanned roots for selfhost:env-reference (same treatment as the existing QDRANT_DIM, which is separately documented where it's read inside src/selfhost/**).

[superagent-security]

Superagent didn't find any vulnerabilities or security issues in this PR.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.95%. Comparing base (4d0aef1) to head (e22b95d).
⚠️ Report is 1 commits behind head on main.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #4350   +/-   ##
=======================================
  Coverage   93.95%   93.95%           
=======================================
  Files         399      399           
  Lines       36782    36786    +4     
  Branches    13436    13440    +4     
=======================================
+ Hits        34557    34561    +4     
  Misses       1569     1569           
  Partials      656      656           

Files with missing lines	Coverage Δ
src/review/adapters.ts	94.44% <100.00%> (+0.32%)	⬆️
src/review/rag.ts	100.00% <100.00%> (ø)

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.