feat: Add extraction of default and descriptions for fields in Pydantic models

[Mahhheshh (Mahhheshh)]

fixes #31

Adds:

• test_parameters.py to test_core
• _extract_pydantic_fields method in py/src/braintrustdata/parameters.py file.

Question:
I should have asked this before; will there be multiple Pydantic fields for a Single Model? How do we need to handle that.

closure:
AI helped me with the tests

[Abhijeet Prasad (AbhiPrasad)]

a single Pydantic model can have multiple fields. I'm not sure if you should keep metadata per field under schema.properties or flatten into multiple eval params. Can you try both and see what it looks like in the UI?

[Abhijeet Prasad (AbhiPrasad)]

Also I finally fixed CI for fork PRs. Mind rebasing this PR?

[Mahhheshh (Mahhheshh)]

Apologies for the delay;

The description and defaults are now an array of key-value pairs,
where the key is the field name and the value is the default value/description.

I’m not great at UI, but the frontend could check if it’s a string; if it is, it could display as it did previously. If it's an array, perhaps under the EvalParams, we could have a UL where each field name shows its respective value.

also it would be great to hear from the PM/designers before merging this in.

and; AI helped me with tests

Bonus

[Abhijeet Prasad (AbhiPrasad)]

let me do some UI tests with this and adjust stuff accordingly.

[Mahhheshh (Mahhheshh)]

Returns scalar value if the model has single field, if has multiple fields then flattens the values in dicts.

compatible with validate_parameters()