feat: expose model_version in responses for humanitarian, anonymize, and ocr schemas

[Opulencechuks]

Closes #272
Summary

This PR exposes the model_version in the response payloads for the Humanitarian verification, PII anonymization, and OCR endpoints. Prior to this change, consumers of these APIs could not determine which model version (e.g. gpt-4o-mini or llama-3.3-70b-versatile) processed their request.

model_version is resolved dynamically at runtime from settings (settings.openai_model or settings.groq_model) based on the active or resolved provider.

Testing

The changes were verified using both schema-level unit tests and route integration tests.

Tests Run

We added new snapshot tests to pin the model_version field in the expected output structure, and updated the integration test assertions to verify the value returned on the endpoints.

# To run the test suite locally once the Python environment is set up:
pytest

Screenshots

N/A (API changes only)

Checklist

• Tests added or updated for new/changed behavior
• Existing tests pass
• No secrets, keys, or seed phrases committed
• Follows the coding conventions in the relevant service README
• PR is focused on a single concern

[kilodesodiq-arch]

Awesome work on this! Thanks for the contribution 🚀 Merging now.

[kilodesodiq-arch]

Hi 👋 Thanks for this PR! There are some merge conflicts with the current main. Could you rebase onto main and push when you have a moment? Once it's clean we'll get it merged 🚀