[Misc] Remove LoRA log (#15388)

Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>

vllm/config.py

@@ -2373,12 +2373,6 @@ class LoRAConfig:
             self.lora_dtype = model_config.dtype
         elif isinstance(self.lora_dtype, str):
             self.lora_dtype = getattr(torch, self.lora_dtype)
-        if model_config.quantization and model_config.quantization not in [
-                "awq", "gptq"
-        ]:
-            # TODO support marlin
-            logger.warning("%s quantization is not tested with LoRA yet.",
-                           model_config.quantization)
 
     def verify_with_scheduler_config(self, scheduler_config: SchedulerConfig):
         # Reminder: Please update docs/source/features/compatibility_matrix.md

vllm/lora/punica_wrapper/punica_gpu.py

@@ -78,10 +78,6 @@ class PunicaWrapperGPU(PunicaWrapperBase):
                                          ...], scale: float, **kwargs):
         """
         Performs GEMM  for multiple slices of lora_a.
-        When `is_prefill is` true, it indicates that it is currently the
-        prefill stage, and the `_shrink_prefill` function should be called.
-        Otherwise, it is the decode stage, and the _shrink_decode function
-        should be called.
             
         Semantics:
         for i in range(len(lora_a_stacked)):