docs: update README to reflect ChromaDB as the active vector store

[divyalankeshwar2007-svg]

📋 Summary

This PR aligns the README documentation with the current vector database implementation used in the project.

---

🔄 Changes Made

• Replaced outdated Pinecone references with ChromaDB references
• Updated vector store descriptions to reflect the active architecture
• Updated retrieval workflow documentation for consistency
• Removed Pinecone-specific setup instructions from prerequisites
• Improved overall documentation clarity for contributors

---

❓ Why this change is needed

The backend has migrated to ChromaDB as the active vector store. However, the README still contained legacy Pinecone references. This could confuse contributors during setup and onboarding by suggesting a deprecated or unsupported vector database.

---

🔗 Related Issue

Closes #555

---

👍 Final Note

This update ensures that the documentation accurately reflects the current architecture and improves onboarding clarity for new contributors.

[divyalankeshwar2007-svg]

Hi @param20h ,

I checked the failing CI job and the failure appears to be unrelated to the documentation changes in this PR.

The backend lint/import check is failing with:

ModuleNotFoundError: No module named 'pymupdf4llm'

during application imports in the test environment.

Since this PR only updates README documentation to reflect ChromaDB usage, it seems the failure is caused by a missing backend dependency in the CI environment rather than the changes introduced here.

Could someone please take a look or re-run the workflow once the dependency issue is resolved?

Thank you.