Fix warmup error when pooling (embedding and clasification) is running (HabanaAI#1442)

by checking if pooling_param is None

@czhu15  @yangulei @ranzhejiang

Author: testdig

vllm/model_executor/layers/pooler.py

@@ -241,6 +241,7 @@ def forward(self, pooled_data: Union[list[torch.Tensor], torch.Tensor],
         dimensions_list = [
             pooling_param.dimensions
             for _, pooling_param in pooling_metadata.seq_groups
+            if pooling_param is not None
         ]
         if any(d is not None for d in dimensions_list):
             # change the output dimension