[V0 Deprecation] Remove V0 sampling metadata (#25345)

Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>

vllm/model_executor/models/deepseek_vl2.py

@@ -15,7 +15,6 @@ from transformers import BatchFeature
 
 from vllm.config import VllmConfig
 from vllm.distributed import get_tensor_model_parallel_world_size
-from vllm.model_executor import SamplingMetadata
 from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.model_executor.model_loader.utils import set_default_torch_dtype
 from vllm.model_executor.models.transformers import replace_linear_class
@@ -647,10 +646,8 @@ class DeepseekVLV2ForCausalLM(nn.Module, SupportsMultiModal, SupportsPP):
     def compute_logits(
         self,
         hidden_states: torch.Tensor,
-        sampling_metadata: SamplingMetadata,
     ) -> Optional[torch.Tensor]:
-        return self.language_model.compute_logits(hidden_states,
-                                                  sampling_metadata)
+        return self.language_model.compute_logits(hidden_states)
 
     def load_weights(self, weights: Iterable[tuple[str,
                                                    torch.Tensor]]) -> set[str]: