Add authentication setup and configuration details to README; create secrets.toml and .env.example files

[JisanAR03]

This pull request introduces authentication setup, API key configuration, and significant updates to the application's logic, including fallback behavior for the get_ragchain function. The most important changes include adding configuration instructions for Auth0 and API keys, replacing deprecated Streamlit methods, and enhancing the retrieval-augmented generation (RAG) chain with a fallback to an LLM-only approach.

Authentication and Configuration:

• Added a new "Configuration" section in the README.md with instructions for setting up Auth0 credentials in secrets.toml and API keys in .env for MistralAI and Hugging Face.
• Introduced example configuration files: .env.example for API keys and secrets.toml for Auth0 settings. [1] [2]

Streamlit Authentication Updates:

• Updated src/mvt/app.py to replace deprecated st.experimental_user with st.user for authentication checks and user data handling, in accordance with Streamlit's deprecation notice effective May 11, 2025.[1] [2]

Retrieval-Augmented Generation (RAG) Enhancements:

• Updated get_ragchain in src/mvt/main.py to include a fallback mechanism. If the FAISS index fails to load, the function now falls back to an LLM-only approach, ensuring robustness.
• Moved the definition of the LLM and prompt string earlier in the get_ragchain function for better readability and logical flow. [1] [2]

@gcapuzzi Could you please review this PR? I’d appreciate any suggestions or feedback you may have.

[gcapuzzi]

Hi there, thanks so much for your pull request! We really appreciate you taking the time to contribute. Just a heads-up that we can't directly accept pull requests onto the mvt-streamlit branch since it's our production code. Let us know if you have any questions

[JisanAR03]

@gcapuzzi Thank you for your response. If possible, I'd like to communicate with you on Slack as well, or we can definitely discuss this further in next meeting.

[RAWx18]

Hi there, thanks so much for your pull request! We really appreciate you taking the time to contribute. Just a heads-up that we can't directly accept pull requests onto the mvt-streamlit branch since it's our production code. Let us know if you have any questions

I've opened this PR (#87) to contribute to the agents branch as part of the LFX Mentorship program. I'd appreciate your feedback on the current implementation. Please let me know if there are any improvements or additional changes you'd like me to make—whether it's code refinement, test coverage, documentation, or anything else that could enhance the contribution.

Looking forward to your suggestions!