Add seed parameter (optional) and custom evaluation metric for citations overlap

[pamelafox]

Purpose

This pull request introduces several changes to add a new metric for evaluating citation matches and to incorporate a random seed for reproducibility in various functions. The most important changes include adding the new CitationsMatchedMetric class, updating configuration files to include the new metric, and modifying functions to accept and use a seed parameter.

New Metric Addition:

• evals/eval_config.json: Updated the requested_metrics to include "citations_matched" and added a seed parameter in target_parameters.
• evals/evaluate.py: Added the CitationsMatchedMetric class to compute the percentage of citations matched between the response and the ground truth. Registered this new metric in the evaluation pipeline. [1] [2]

Seed Parameter for Reproducibility:

• src/backend/fastapi_app/api_models.py: Added a seed field to the ChatRequestOverrides model.
• src/backend/fastapi_app/rag_advanced.py: Modified multiple functions (generate_search_query, prepare_context, answer) to accept and use the seed parameter. [1] [2] [3] [4]
• src/backend/fastapi_app/rag_base.py: Updated the get_params method to include the seed parameter.
• src/backend/fastapi_app/rag_simple.py: Added the seed parameter to the answer and answer_stream methods. [1] [2]

Does this introduce a breaking change?

When developers merge from main and run the server, azd up, or azd deploy, will this produce an error?
If you're not sure, try it out on an old environment.

[ ] Yes
[X] No

Type of change

[ ] Bugfix
[X] Feature
[ ] Code style update (formatting, local variables)
[ ] Refactoring (no functional changes, no api changes)
[ ] Documentation content changes
[ ] Other... Please describe:

Code quality checklist

See CONTRIBUTING.md for more details.

• The current tests all pass (python -m pytest).
• I added tests that prove my fix is effective or that my feature works
• I ran python -m pytest --cov to verify 100% coverage of added lines
• I ran python -m mypy to check for type errors
• I either used the pre-commit hooks or ran ruff manually on my code.