feat: inference API cleanup

[ishaansehgal99]

This PR does some cleanup on the inference API. Introduces:

1. ModelConfig Dataclass
2. Allow additional user args
3. GenerateKwargs Model
4. More descriptive comments/metadata for model fields
5. Rename inference-api.py -> inference_api.py - needed for naming convention/importability
6. Introduce tests for inference_api.py

This PR requires bumping up images and having Kaito use updated image tag (in this case 0.0.2).

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (7f50583) 65.27% compared to head (db7e010) 65.27%.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #233   +/-   ##
=======================================
  Coverage   65.27%   65.27%           
=======================================
  Files          14       14           
  Lines        1123     1123           
=======================================
  Hits          733      733           
  Misses        352      352           
  Partials       38       38           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.