### Check for existing issues

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

### What happened?

### What happened?

When using `reasoning_effort` with a Bedrock-hosted model that is **not** Anthropic, Amazon Nova 2, or a `gpt-oss` model (e.g. Qwen3 models available on Bedrock), the parameter is silently dropped with no error and no effect on the response — the model never returns reasoning/thinking content.

Root cause: `AmazonConverseConfig._handle_reasoning_effort_parameter()` in `litellm/llms/bedrock/chat/converse_transformation.py` branches on model name:

```python def _handle_reasoning_effort_parameter(self, model: str, reasoning_effort: str, optional_params: dict) -> None: if "gpt-oss" in model: optional_params["reasoning_effort"] = reasoning_effort elif self._is_nova_2_model(model): reasoning_config = self._transform_reasoning_effort_to_reasoning_config(reasoning_effort) optional_params.update(reasoning_config) else: mapped_thinking = AnthropicConfig._map_reasoning_effort( reasoning_effort=reasoning_effort, model=model, custom_llm_provider="bedrock", llm_provider=…