### What happened?

When the litellm proxy detects the issue "litellm.InternalServerError: The model is repeating the same chunk" it seems to call the custom logging callback thousands of times (and floods our telemetry system). Is it possible that the safety checker does not actually terminate the processing of the response?

### Relevant log output

```shell Traceback (most recent call last):\n File \"/usr/local/lib/python3.13/site-packages/litellm/litellm_core_utils/streaming_handler.py\", line 1708, in __anext__\n processed_chunk: Optional[ModelResponseStream] = self.chunk_creator(\n ~~~~~~~~~~~~~~~~~~^\n chunk=chunk\n ^^^^^^^^^^^\n )\n ^\n File \"/usr/local/lib/python3.13/site-packages/litellm/litellm_core_utils/streaming_handler.py\", line 1430, in chunk_creator\n raise exception_type(\n ...<3 lines>...\n )\n File \"/usr/local/lib/python3.13/site-packages/litellm/litellm_core_utils/streaming_handler.py\", line 1419, in chunk_creator\n return self.return_processed_chunk_logic(\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^\n completion_obj=completion_obj,\n ^^^^^^^^^^^^^^^^^…