### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

LiteLLM currently incorrectly registers GPT-5.5 and GPT-5.6 variants (such as gpt-5.5, gpt-5.5-2026-04-23, gpt-5.6-luna, gpt-5.6-sol, gpt-5.6-terra) as supporting reasoning_effort="none".

Because LiteLLM thinks these models support none effort, it bypasses its local parameter validation for temperature and allows values like 0.0 or 0.5 to pass through. However, because these are pure reasoning models that do not support none effort, the upstream OpenAI API rejects the request with an HTTP 400 (Unsupported value: 'temperature' does not support 0 with this model).

LiteLLM should catch this locally and raise litellm.utils.UnsupportedParamsError (as it already correctly does for gpt-5-mini and o1).

```python import os import litellm

os.environ["OPENAI_API_KEY"] = "your-api-key"

try: litellm.completion( model="openai/gpt-5-mini", messages=[{"role": "user", "content": "Hi"}], temperature=0.0 ) except litellm.utils.UnsupportedParamsError as e: print("Caught expected local validation error"…