feat: add AI research assistant suite module

[TuncayEnsi]

@algora-pbc /claim #16

Closes #16

What this adds

Adds a self-contained ai-research-assistant-suite/ module for automated peer review reports, reproducibility checks, and research gap discovery.

Requirement mapping

• Auto peer review reports -> implemented in ai-research-assistant-suite/src/lib/peer-review.ts; includes domain templates, review-dataset profile metadata, clarity/statistics/ethics/data availability/citation/claims-versus-evidence checks, and schema-valid findings.
• Project/manuscript analysis -> implemented against ResearchProject sections and artifacts in src/types.ts and src/data/demo-data.ts.
• Reproducibility checker -> implemented in src/lib/reproducibility.ts; checks notebooks, pinned environments, raw data manifests, pipeline artifacts, validation splits, deterministic seeds, expected outputs, sandbox metadata, confidence score, and previous attempts.
• Research gap finder -> implemented in src/lib/gap-finder.ts; uses project keywords, citation count, replication count, unresolved questions, limitations, and negative-result follow-ups from the seed corpus.
• Combined dashboard and reviewer smoke-test APIs -> implemented in src/server.ts.
• Tests -> implemented in tests/assistant-suite.test.ts.

How to verify in under 5 minutes

cd ai-research-assistant-suite
npm install
npm run typecheck
npm test
npm run build
npm run dev

Open http://localhost:3002 and verify the dashboard shows Auto peer review, Reproducibility checker, Research gap finder, review profile metadata, sandbox runner metadata, and gap signals.

Optional API smoke tests:

curl -s http://localhost:3002/api/dashboard/project_organoid_response | jq .
curl -s http://localhost:3002/api/review/project_organoid_response | jq .
curl -s http://localhost:3002/api/reproducibility/project_organoid_response | jq .
curl -s http://localhost:3002/api/gaps/project_organoid_response | jq .

Validation

Commands run:

• npm run typecheck
• npm test - 1 file, 4 tests
• npm run build
• npm audit --omit=dev - 0 vulnerabilities
• Browser demo verified locally: updated dashboard renders review model metadata, sandbox runner metadata, previous-attempt count, low-replication gap signals, and negative-result follow-up signals without console errors.

Demo

• Short demo video: ai-research-assistant-suite-demo.mp4
• Screenshot: dashboard.png
• Both files are committed in this PR under ai-research-assistant-suite/docs/demo/.

Scope

This PR is intentionally self-contained under ai-research-assistant-suite/ and avoids unrelated root-level changes.

[TuncayEnsi]

/claim #16

[TuncayEnsi]

@algora-pbc /claim #16