[Model] Support pooling models (#3122)

### What this PR does / why we need it?

Support pooling models (like `bge-reranker-v2-m3`) in vllm-ascend, this
pr covered the three model types of embed (cls_token, mean_token,
lasttoken).

After this
[commit](17373dcd93),
vllm has provided support for adapting pooling models on the v1 engine.
This PR includes corresponding adaptations on the vllm-ascend side.

Fixes #1960

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: lianyibo <lianyibo1@kunlunit.com>
Signed-off-by: MengqingCao <cmq0113@163.com>
Co-authored-by: MengqingCao <cmq0113@163.com>

vllm_ascend/attention/attention_mask.py

@@ -35,7 +35,6 @@ class AttentionMaskBuilder:
         self.attn_mask_cache = None
         self._seq_len_cached = 0
         self.device = device
-        self.pooling_mask = None
         self.mla_mask = None
         self.chunked_prefill_attn_mask = None
         self.pcp_mla_mask = None
@@ -50,14 +49,6 @@ class AttentionMaskBuilder:
         return self.attn_mask_cache[:max_seq_len, :max_seq_len].contiguous(
         ).to(self.device, non_blocking=True)
 
-    def get_pooling_mask(self):
-        if self.pooling_mask is None:
-            # the compressed attention mask for npu_fusion_attention sparse mode 4
-            self.pooling_mask = torch.triu(torch.ones(
-                2048, 2048), diagonal=1).to(torch.bool).to(self.device,
-                                                           non_blocking=True)
-        return self.pooling_mask
-
     def get_splitfuse_attn_mask(self) -> torch.Tensor:
         if self.chunked_prefill_attn_mask is None:
             self.chunked_prefill_attn_mask = torch.triu(