[Feature] optimize sp & qwen3 next support sp. (#3225)

This PR will accomplish the following tasks: 
**optimize SP**
In the old version implementation, the first layer was all_reduce, which
used rms to split chunks. We changed it to perform reduce_scatter on the
embedding side, replace one all_reduce operation and one chunk with one
reduce_scatter operation.
**Support qwen3 next**
Since Qwen3 Next includes a linear attention module, the prefix name of
this module cannot take effect directly.


- vLLM version: v0.11.0rc3
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

---------

Signed-off-by: weijinqian_v1 <weijinqian@huawei.com>
Co-authored-by: weijinqian_v1 <weijinqian@huawei.com>

vllm_ascend/ops/vocab_parallel_embedding.py

@@ -20,7 +20,7 @@ from typing import Optional, Tuple
 import torch
 from torch import nn
 from torch.nn.parameter import Parameter
-from vllm.distributed import divide, tensor_model_parallel_all_reduce
+from vllm.distributed import divide
 from vllm.distributed.parallel_state import get_tp_group
 from vllm.model_executor.layers.logits_processor import LogitsProcessor
 from vllm.model_executor.layers.quantization.base_config import (
@@ -163,7 +163,7 @@ class AscendVocabParallelEmbedding(VocabParallelEmbedding):
         if self.tp_size > 1:
             output_parallel.masked_fill_(input_mask.unsqueeze(-1), 0)
         # Reduce across all the model parallel GPUs.
-        output = tensor_model_parallel_all_reduce(output_parallel)
+        output = torch.ops.vllm.maybe_pad_and_reduce(output_parallel)
         return output