[refactor] refactor model runner capture model (#5230)

### What this PR does / why we need it?
Refactor the `capture_model` method in model_runner to directly reuse
the method from vLLM.

Currently, most of the logic in the capture_model method is similar to
that in the vllm code. Directly using the vllm method can reduce the
maintenance cost of the vllm-ascend code. Modify as follows:
1、refactor capture_model function, directly inheriting community methods
2、refactor initialize_aclgraph_capture function, move to
initialize_attn_backend

### Does this PR introduce _any_ user-facing change?
No

### How was this patch tested?

- vLLM version: release/v0.13.0
- vLLM main:
ad32e3e19c

Signed-off-by: weiguihua2 <weiguihua2@huawei.com>
Co-authored-by: weijinqian0 <1184188277@qq.com>

vllm_ascend/attention/mla_cp.py

@@ -1,4 +1,4 @@
-from typing import ClassVar, Optional, Tuple, TypeVar
+from typing import Optional, Tuple, TypeVar
 
 import numpy as np
 import torch
@@ -12,7 +12,7 @@ from vllm.distributed import (get_dcp_group,
 from vllm.forward_context import ForwardContext, get_forward_context
 from vllm.utils.math_utils import cdiv
 from vllm.v1.attention.backends.utils import AttentionCGSupport
-from vllm.v1.kv_cache_interface import MLAAttentionSpec
+from vllm.v1.kv_cache_interface import AttentionSpec, MLAAttentionSpec
 
 # isort: off
 from vllm_ascend.attention.mla_v1 import (AscendMLADecodeMetadata,
@@ -37,9 +37,6 @@ M = TypeVar("M", bound=AscendMLAMetadata)
 
 
 class AscendMlaCPMetadataBuilder(AscendMLAMetadataBuilder):
-    # Does this backend/builder support ACL Graphs for attention (default: no).
-    aclgraph_support: ClassVar[AttentionCGSupport] = \
-    AttentionCGSupport.UNIFORM_BATCH
     """
     NOTE: Please read the comment at the top of the file before trying to
     understand this class
@@ -74,6 +71,16 @@ class AscendMlaCPMetadataBuilder(AscendMLAMetadataBuilder):
                                               dtype=torch.uint8,
                                               device=device)
 
+    @classmethod
+    def get_cudagraph_support(
+        cls: type["AscendMlaCPMetadataBuilder"],
+        vllm_config: VllmConfig,
+        kv_cache_spec: AttentionSpec,
+    ) -> AttentionCGSupport:
+        # Explicit override in case the underlying builder specialized this getter.
+        # @override omitted only because of mypy limitation due to type variable.
+        return AttentionCGSupport.UNIFORM_BATCH
+
     def set_num_actual_tokens(
         self,
         common_attn_metadata: AscendCommonAttentionMetadata,