### Check for existing issues

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

### What happened?

Disclosure: found this while troubleshoot using AI.

related to #31610, in caching/valkey_semantic_cache.py 2 calls of ``` embedding = await self._get_async_embedding(prompt, **kwargs) ``` in lines 282 and 301 are not sending expected `metadata` but **kwargs

supposedly it should be something similar to

``` embedding = await self._get_async_embedding( prompt, metadata=kwargs.get("metadata"), ) ```

workaround applied for caching to work: patched caching/caching.py like #31610 patched caching/valkey_semantic_cache.py (attached) [valkey_semantic_cache.py](https://github.com/user-attachments/files/29734382/valkey_semantic_cache.py)

### Steps to Reproduce

1. docker pull litellm:v1.92.0-rc1 2. setup valkey semantic cache in config.yaml with proper embedding 3. monkey patch the cache not shown issue on litellm/_logging.py #32318 [_logging.py](https://github.com/user-attachments/files/29734028/_logging.py) 4. send 2 indentical chat_completion via curl 5. observe the log for errors

### Relevant log output

```shell 04:18:06 - LiteLLM:DE…