Implement session-based FAISS indexing and update server/frontend for session handling

[DivyaSreenijaKasa]

Summary

Implemented session-based FAISS indexing to isolate uploaded PDFs.
Each upload now generates a unique session_id and stores its FAISS index
in a dedicated folder. This prevents cross-document mixing and ensures
queries return results only from the correct session.

Related issue

Closes #56

Testing

• Verified app starts successfully
• Uploaded multiple PDFs and confirmed unique session IDs are generated
• Queried each session independently and validated answers come only from the correct document
• Frontend and server updated to pass session_id correctly

Screenshots / recordings

Swagger UI and local tests confirm correct isolation of sessions.

Notes

• Updated backend (rag-service/main.py), server (server.js), and frontend (src/index.js)
• Dependency files (package.json, package-lock.json) updated
• rag-service/faiss_store/ contains runtime indexes and should remain ignored in version control

Future work

• Add cleanup logic to remove expired or unused FAISS indexes to prevent disk bloat
• Update README.md to document persistence behavior and usage instructions
• Explore support for multi-document history or named sessions

[FireFistisDead]

@DivyaSreenijaKasa resolve the merge conflicts

[DivyaSreenijaKasa]

@FireFistisDead Thanks for pointing that out. I’ll resolve the merge conflicts and update the PR shortly.

[DivyaSreenijaKasa]

@FireFistisDead Conflicts have been resolved and the branch is up to date.
This PR is ready for review and merge.
If there are any further changes needed, please let me know.