[Perf] Add new npu_fused_infer_attention_score op to improve perfomance in splitfuse cases and resolve long-seq mask problems (#2962)

### What this PR does / why we need it?
Add new npu_fused_infer_attention_score op to improve perfomance in
splitfuse cases and resolve long-seq mask problems .

1. The original op's performance is suboptimal in certain scenarios,
necessitating optimization through the _new op_
(npu_fused_infer_attention_score)。
2. For ultra-long sequences (128k), the original operator will allocate
a large attn_mask, which consumes excessive CPU memory. In contrast, the
_new op_ supports a fixed-size compressed mask, effectively resolving
this issue.

NOTE1: The current PR retains the original logic and uses a version
check of the CANN package to determine whether the _new op_ can be
enabled. This ensures no impact on existing users. In future versions,
this version check and the original logic will be deprecated, and the
_new op_ scheduling will be uniformly adopted.
NOTE2: This pr relies on future CANN version, which is not available
now.
NOTE3: To enable the new op in chunked prefill, the parameter
additional_config should be set like `--additional-config
'{"ascend_scheduler_config":
{"enabled":true,"enable_chunked_prefill":true}}' \` at least.

### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
CI passed




- vLLM version: v0.10.2
- vLLM main:
6c5f82e5aa

---------

Signed-off-by: tangtianyi <tangtianyi4@huawei.com>
Signed-off-by: Angazenn <supperccell@163.com>
Co-authored-by: Angazenn <supperccell@163.com>

vllm_ascend/worker/model_runner_v1.py

@@ -301,8 +301,13 @@ class NPUModelRunner(LoRAModelRunnerMixin):
             use_mla=self.model_config.use_mla,
         )
 
-        self.attn_mask_builder = AttentionMaskBuilder(
-            self.model_config.max_model_len, self.dtype)
+        if torch.version.cann.startswith("8.3"):
+            self.attn_mask_builder = AttentionMaskBuilder(
+                self.scheduler_config.max_num_batched_tokens, self.dtype,
+                self.device)
+        else:
+            self.attn_mask_builder = AttentionMaskBuilder(
+                self.model_config.max_model_len, self.dtype)
 
         # Set up speculative decoding.
         self.spec_attn_mask = None
@@ -860,8 +865,11 @@ class NPUModelRunner(LoRAModelRunnerMixin):
                              attn_state) -> torch.Tensor:
         # Chunk Prefill situation.
         if attn_state == AscendAttentionState.ChunkedPrefill and not self.vllm_config.model_config.use_mla:
-            return self.attn_mask_builder.get_splitfuse_attn_mask(
-                seq_lens, position, self.dtype, self.device)
+            if torch.version.cann.startswith("8.3"):
+                return self.attn_mask_builder.get_splitfuse_attn_mask()
+            else:
+                return self.attn_mask_builder.get_splitfuse_attn_mask(
+                    seq_lens, position, self.dtype, self.device)
         # Prefill without cache situation.
         elif attn_state == AscendAttentionState.PrefillNoCache:
             max_seq_len = max(seq_lens, default=0)