Make RAG document endpoints deployable

[ar-amk]

Part of the open Algora bounty for [ISAAC-497] Implement an enhanced RAG Pipeline for Scientific/Research Workflows.

/claim #45

Bounty reference: https://algora.io/isaac/bounties/clq18zr98000ejs0gt0nv7gwu

Summary

• Build the internal /api/fetch-documents URL from the incoming RAG request origin instead of hardcoding localhost:3000.
• Use CHROMA_PATH with the existing Docker default for document fetches.
• Reject unsupported methods and missing query input before hitting Chroma.
• Preserve the resolved RAG chat temperature when streaming answers, instead of falling back to the old hardcoded 0 while touching this flow.
• Add focused helper coverage for Chroma path and same-origin URL resolution.

Why this helps the scientific RAG workflow

The RAG chat path should work the same locally, in Docker, and behind a deployed host. Hardcoded localhost routing and inconsistent Chroma host defaults make the document retrieval path fragile once the app is deployed. This PR keeps the change focused on deployability, so uploaded research documents can be retrieved reliably by the chat endpoint.

Verification

From ui/:

npx vitest run __tests__/rag-documents.test.ts --reporter verbose
npx prettier --check pages/api/rag-chat.ts pages/api/fetch-documents.ts utils/server/rag-documents.ts __tests__/rag-documents.test.ts
npx tsc --noEmit --pretty false
npm run lint -- --file pages/api/rag-chat.ts --file pages/api/fetch-documents.ts --file utils/server/rag-documents.ts --file __tests__/rag-documents.test.ts
git diff --check

Results:

• Targeted Vitest passed: 3 tests
• Prettier passed
• TypeScript passed
• ESLint passed
• Diff whitespace check passed