fix: support max reasoning and custom model selection for 1.0.0

tests/unit/test_openai_responses.py

@@ -85,6 +85,15 @@ async def test_text_and_image_use_responses_transport(transport_model):
     assert result.usage_metadata["total_tokens"] == 13
 
 
+@pytest.mark.asyncio
+async def test_max_reasoning_is_sent_unchanged_to_responses(transport_model):
+    model, requests, _ = transport_model
+    model.reasoning_effort = "max"
+    await model.ainvoke([HumanMessage(content="Check")])
+    assert requests[0].url.path == "/v1/responses"
+    assert json.loads(requests[0].content)["reasoning"]["effort"] == "max"
+
+
 @pytest.mark.asyncio
 async def test_function_call_roundtrip_uses_call_id(transport_model):
     model, requests, outputs = transport_model