Enable kvcache_nz for the decode process in torchair graph mode (#1098)

What this PR does / why we need it?
Enable kvcache_nz for the decode process in torchair graph mode, which
reduces the time consumed by FA in long sequences.

Does this PR introduce any user-facing change?
If need to enable kvcache_nz, should set the
additional_config.torchair_graph_config.enable_kv_nz=True

How was this patch tested?
1. Tested in deepseek model:
with batchsize 64 and seq_len 1k+3k, 61 layers FA total time improves
20.80ms -> 19.76ms
2. operator precision test: 

[aclnnFusedInferAttentionScoreV3_result.csv](https://github.com/user-attachments/files/20664138/aclnnFusedInferAttentionScoreV3_result.csv)
3. tpot test from @ttanzhiqiang, and curl one result is normal

https://github.com/vllm-project/vllm-ascend/pull/1098#issuecomment-2948542159

https://github.com/vllm-project/vllm-ascend/pull/1098#issuecomment-2954496588

---------

Signed-off-by: chenwaner <861645847@qq.com>

vllm_ascend/ascend_config.py

@@ -58,6 +58,7 @@ class TorchairGraphConfig:
             "enable_multistream_moe", False)
         self.enable_view_optimize = torchair_graph_config.get(
             "enable_view_optimize", True)
+        self.enable_kv_nz = torchair_graph_config.get("enable_kv_nz", False)
 
         if not isinstance(self.graph_batch_sizes, list):
             raise TypeError("graph_batch_sizes must be list[int]")