Please add an option that lets us use "max_completion_tokens" instead of "max_tokens" #908

Describe the bug
问题描述
A clear and concise description of what the bug is.

GPT-5/5.1 uses “max_completion_tokens” instead of “max_tokens”, so an error occurs when using the old parameter

To Reproduce
如何复现
Steps to reproduce the behavior:

• Simply send any request using the GPT-5/5.1 model

Expected behavior
期望行为
A clear and concise description of what you expected to happen.

• It should work the same way as the legacy models

Screenshots
截图说明
If applicable, add screenshots to help explain your problem.

Please complete the following information):
请补全以下内容

• OS: [e.g. Windows] Windows 11
• Browser: [e.g. chrome, safari] Chrome
• Extension Version: [e.g. v2.0.2] v2.5.9

Additional context
其他
Add any other context about the problem here.