[Bug]: [autogen_ext] "ValueError: No stop reason found" always raised when llm usage returned

[nomagicln]

Describe the bug
When executing the sample code, this error is always thrown: "ValueError: No stop reason found".

Steps to reproduce

import asyncio

from autogen_ext.models.openai import OpenAIChatCompletionClient
from autogen_core.models import UserMessage, ModelInfo, ModelFamily

client = OpenAIChatCompletionClient(
    model="deepseek-chat",
    base_url="https://api.deepseek.com/v1",
    api_key="sk-xxxxx",
    model_info=ModelInfo(
        vision=False,
        function_calling=False,
        json_output=False,
        family=ModelFamily.UNKNOWN,
    ),
)


async def main():
    messages = [
        UserMessage(content="OUTPUT 1 only", source="user"),
    ]

    # Create a stream.
    stream = client.create_stream(messages=messages)

    # Iterate over the stream and print the responses.
    print("Streamed responses:")
    async for response in stream:  # type: ignore
        if isinstance(response, str):
            # A partial response is a string.
            print(response, flush=True, end="")
        else:
            # The last response is a CreateResult object with the complete message.
            print("\n\n------------\n")
            print("The complete response:", flush=True)
            print(response.content, flush=True)
            print("\n\n------------\n")
            print("The token usage was:", flush=True)
            print(response.usage, flush=True)


if __name__ == "__main__":
    asyncio.run(main())

Model Used
deepseek-chat

Expected Behavior
execute normally

Screenshots and logs
No response

Additional Information

I found the place where the problem occurs. According to the code logic here, only when the usage is None and the stop_reason is also None will the finish_reason be set to the stop_reason. It's very strange why this is done.

https://github.com/microsoft/autogen/blame/8a83262a905675f15dfca387ba4a8de7b6cf0635/python/packages/autogen-ext/src/autogen_ext/models/openai/_openai_client.py#L695

stop_reason = None
# ...
stop_reason = choice.finish_reason if chunk.usage is None and stop_reason is None else stop_reason