fix: o1 model error, use max_completion_tokens instead of max_tokens.

[jiangbo721]

Summary

fix: o1 model error, when the max_tokens switch is on, a parameter error is indicated .use max_completion_tokens instead of max_tokens.

Tip

Close issue syntax: Fixes #<issue number> or Resolves #<issue number>, see documentation for more details.

Screenshots

{"event": "error", "conversation_id": "3106d28d-5807-43e5-8a18-107fbd5f623c", "message_id": "4fc7a62a-9b6a-486f-85d5-9bb4f4ee97f6", "created_at": 1735021270, "code": "completion_request_error", "status": 400, "message": "[azure_openai:llm:o1-preview] Bad Request Error, Error code: 400 - {'error': {'message': "Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead.", 'type': 'invalid_request_error', 'param': 'max_tokens', 'code': 'unsupported_parameter'}}"}

Before	After
...	...

Checklist

Important

Please review the checklist below before submitting your pull request.

• This change requires a documentation update, included: Dify Document
• I understand that this PR may be closed in case there was no previous discussion or issues. (This doesn't apply to typos!)
• I've added a test for each change that was introduced, and I tried as much as possible to make a single atomic change.
• I've updated the documentation accordingly.
• I ran dev/reformat(backend) and cd web && npx lint-staged(frontend) to appease the lint gods

[jiangbo721]

I replaced the code in azure_openai with the same judgement method as openai. Use model.startswith(‘o1’) instead of if ‘o1’ in model