feat(integrations): openai-agents: add usage and response model reporting for chat and invoke_agent spans

[constantinius]

Issues

Closes https://linear.app/getsentry/issue/TET-1457/py-openai-agents-attributes-missing

[linear]

TET-1457 Py OpenAI Agents - attributes missing

[codecov]

Codecov Report

❌ Patch coverage is 85.71429% with 4 lines in your changes missing coverage. Please review.
✅ Project coverage is 83.97%. Comparing base (6c6705a) to head (a34668b).
⚠️ Report is 2 commits behind head on master.
✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
...y_sdk/integrations/openai_agents/patches/models.py	82.35%	0 Missing and 3 partials ⚠️
...k/integrations/openai_agents/spans/invoke_agent.py	66.66%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@ Coverage Diff @@## master #5157 +/- ## ========================================== - Coverage 83.98% 83.97% -0.02%  ========================================== Files 181 181 Lines 18395 18419 +24 Branches 3275 3281 +6 ========================================== + Hits 15449 15467 +18 - Misses 1937 1938 +1 - Partials 1009 1014 +5 

Files with missing lines	Coverage Δ
...dk/integrations/openai_agents/patches/agent_run.py	83.33% <100.00%> (+0.52%)	⬆️
...y_sdk/integrations/openai_agents/patches/runner.py	95.65% <100.00%> (+0.19%)	⬆️
..._sdk/integrations/openai_agents/spans/ai_client.py	100.00% <100.00%> (ø)
...k/integrations/openai_agents/spans/invoke_agent.py	84.61% <66.66%> (-1.88%)	⬇️
...y_sdk/integrations/openai_agents/patches/models.py	86.84% <82.35%> (-4.07%)	⬇️

... and 2 files with indirect coverage changes

[alexander-alderman-webb]

out of curiosity, why do we need to capture the raw response model?

[constantinius]

This is the only place I found to get the response model. OpenAI Agents does not expose this, we only get the request model, but not the response.