[refactor] refactor excute_model and _dymmy_run method

tests/ut/worker/test_pcp_manager.py

@@ -123,6 +123,7 @@ def test_update_tokens_for_pcp_basic(tokens, num_reqs, num_computed_tokens,
     vllm_config = MagicMock()
     vllm_config.model_config = MagicMock()
     vllm_config.speculative_config.num_speculative_tokens = 0
+    vllm_config.scheduler_config.max_num_seqs = 1000
 
     pcp_manager = PCPManager(pcp_world_size=pcp_size,
                              pcp_rank=0,

vllm_ascend/attention/sfa_v1.py

@@ -170,7 +170,7 @@ def __init__(
                 npu_fused_infer_attention_score TND layout's limit of 16, \
                 got {self.decode_threshold}"
             )
-
+        self.reorder_batch_threshold = self.decode_threshold
         self.attn_mask_builder = AttentionMaskBuilder(self.device)
         self.rope_dim = self.model_config.hf_text_config.qk_rope_head_dim
         self.enable_dsa_cp = enable_dsa_cp()