Fixes #1582: Handling of structured CompletionUsage response values for token usage

[elondemirok]

Summary

This PR resolves a compatibility issue caused by recent changes in the Python SDK. Previously, token usage data (completion_tokens, prompt_tokens, total_tokens) was returned as integers. Recent API updates introduced structured CompletionUsage objects, resulting in type errors after eval execution.

• I have tested my fix locally and verified it resolves the issue. Screenshot of output attached.

• I have minimized changes to keep the code easy to review and maintain.
• This PR does not introduce new evals or modify existing evaluation logic.

Final checklist 👀

Submission agreement

• I agree that my submission will be made available under an MIT license and complies with OpenAI's usage policies.
• I acknowledge that GPT-4 access will only be granted, if applicable, to the email address used for my merged pull request.
• I understand that opening a PR, even if it meets the requirements above, does not guarantee the PR will be merged nor GPT-4 access be granted.
• I have filled out all required fields of this form
• ⚠️ It's a bug fix; I haven't used Git LFS for the Eval JSON data
• (Ignore if not submitting code) I have run pip install pre-commit; pre-commit install and have verified that mypy, black, isort, autoflake and ruff are running when I commit and push