ollama[patch]: fix model validation, ensure per-call reasoning can be set, tests (#31927)

* update model validation due to change in [Ollama
client](https://github.com/ollama/ollama) - ensure you are running the
latest version (0.9.6) to use `validate_model_on_init`
* add code example and fix formatting for ChatOllama reasoning
* ensure that setting `reasoning` in invocation kwargs overrides
class-level setting
* tests

libs/partners/ollama/tests/unit_tests/test_llms.py

@@ -48,3 +48,24 @@ def test_validate_model_on_init(mock_validate_model: Any) -> None:
     # Test that validate_model is NOT called by default
     OllamaLLM(model=MODEL_NAME)
     mock_validate_model.assert_not_called()
+
+
+def test_reasoning_aggregation() -> None:
+    """Test that reasoning chunks are aggregated into final response."""
+    llm = OllamaLLM(model=MODEL_NAME, reasoning=True)
+    prompts = ["some prompt"]
+    mock_stream = [
+        {"thinking": "I am thinking.", "done": False},
+        {"thinking": " Still thinking.", "done": False},
+        {"response": "Final Answer.", "done": True},
+    ]
+
+    with patch.object(llm, "_create_generate_stream") as mock_stream_method:
+        mock_stream_method.return_value = iter(mock_stream)
+        result = llm.generate(prompts)
+
+    assert result.generations[0][0].generation_info is not None
+    assert (
+        result.generations[0][0].generation_info["thinking"]
+        == "I am thinking. Still thinking."
+    )