fix: make tool call assertions flexible to prevent CI flakes

[iamemilio]

Summary

• Loosens strict len(response.output) == 1 assertions in tool call integration tests to accept one or more function calls (>= 1), since models may legitimately return multiple parallel tool calls for a single-tool prompt
• Updates follow-up turns to respond to all returned tool calls (not just the first), preventing the tool_call_id not responded to API error when the model produces duplicate invocations
• These changes make tests resilient to valid model behavior variations across different providers and model versions without weakening what is actually being tested

Motivation

When evaluating cheaper model alternatives for the Azure CI suite, we observed that some models (e.g. gpt-4.1-nano) occasionally return two parallel get_weather function calls instead of one. This is a valid API response — the model is correctly identifying the tool to call, just being eager about parallelism. The previous == 1 assertions treated this as a failure, creating intermittent CI flakes.

The fix is backwards-compatible: tests still pass when a model returns exactly one call (since 1 >= 1), while also accepting the multi-call case. The core behavior being verified (correct tool invocation, successful follow-up round-trip, final message with text) remains unchanged.

Test plan

• Ran the full responses suite with gpt-4.1-nano via OpenAI — 196 passed, 0 failed, 26 skipped
• Verified the previously flaky tests (test_function_call_output_list_text, test_function_call_output_list_text_multi_block, test_response_non_streaming_custom_tool) pass consistently with the loosened assertions
• Assertions remain logically correct — still verify function_call type, status, name, and successful text response after tool output

[mattf]

can you pass parallel_tool_calls=False to create instead?

it'd be good to have a clear parallel_tool_calls=True test too. historically many models would fail it, but we should behave correctly. especially important is parallel_tool_calls=True and stream=True.