Chat Evaluation Framework

[mcelrath]

This Issue is to track progress on an evaluation framework for "chat" functionality.

• [ x] Extract question/answer pairs from Bitcoin Stack Exchange
• [] Run framework against proposed changes to the RAG, AI engine, or middleware

[staru09]

https://github.com/stanford-futuredata/ARES
https://docs.ragas.io/en/stable/

maybe use this as a starting point.