[Neuron] Add custom_ops for neuron backend (#13246)

Signed-off-by: Liangfu Chen <liangfc@amazon.com>
Co-authored-by: George Novack <gnovack@amazon.com>
Co-authored-by: Aoyu Zhang <aoyuzhan@amazon.com>

vllm/model_executor/layers/logits_processor.py

@@ -53,6 +53,7 @@ class LogitsProcessor(nn.Module):
         # Whether to use gather or all-gather to gather the logits.
         parallel_config = get_current_vllm_config().parallel_config
         self.use_all_gather = current_platform.is_tpu() \
+            or current_platform.is_neuron() \
             or envs.VLLM_USE_V1 \
             or parallel_config.distributed_executor_backend == "external_launcher" # noqa