Describe the bug
Can't use Open AI o1-preview model with ChatCompletionAgent, just like in issue #10272, because it fails with the following exception.

Microsoft.SemanticKernel.HttpOperationException: HTTP 400 (invalid_request_error: unsupported_value)

Microsoft.SemanticKernel.HttpOperationException
HTTP 400 (invalid_request_error: unsupported_value)
Parameter: messages[0].role

Unsupported value: 'messages[0].role' does not support 'system' with this model.

To Reproduce
Steps to reproduce the behavior:

1. Create ChatCompletionAgent from semantic_kernel.agents with o1/o3-mini model deployment:

agent = ChatCompletionAgent(
            service_id=...,
            kernel=...,
            name=...,
            instructions=...,
            description=...,
            arguments=...,
        )

2. Invoke created agent:
   agent.invoke(...)

Expected behavior
It should work, like in corrected .NET version

Screenshots

Platform

• Language: Python
• Source: semantic-kernel==1.20.0, openai==1.66.2
• AI model: OpenAI:GPT o-.. reasoning models
• OS: any

Additional context