fix: preserve content fields in Python serialization

[rksharma-owg]

Fixes #78

Summary

• serialize concrete Content subclasses when Message.model_dump_json() is used
• add a regression test for text message content

Validation

• PYTHONPATH=python uv run --with pydantic --with pytest --python 3.13 --no-project pytest -q tests/test_message_serialization.py (1 passed)
• python3.14 -m compileall -q python/openai_harmony tests/test_message_serialization.py
• git diff origin/main --check

Before this change, the regression test produced content: [{}]; after the change it preserves the text field.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-08T04:59:27.131891Z	308266a	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.