openai[patch]: fix Azure LLM test (#29302)

The tokens I get are:
```
['', '\n\n', 'The', ' sun', ' was', ' setting', ' over', ' the', ' horizon', ',', ' casting', '']
```
so possibly an extra empty token is included in the output.

lmk @efriis if we should look into this further.

Author: ccurme

libs/partners/openai/tests/integration_tests/llms/test_azure.py

@@ -148,7 +148,7 @@ def test_openai_streaming_callback() -> None:
         verbose=True,
     )
     llm.invoke("Write me a sentence with 100 words.")
-    assert callback_handler.llm_streams == 11
+    assert callback_handler.llm_streams == 12
 
 
 @pytest.mark.scheduled
@@ -171,5 +171,5 @@ async def test_openai_async_streaming_callback() -> None:
         verbose=True,
     )
     result = await llm.agenerate(["Write me a sentence with 100 words."])
-    assert callback_handler.llm_streams == 11
+    assert callback_handler.llm_streams == 12
     assert isinstance(result, LLMResult)