[Feature]refactor the npugraph_ex config, support online-infer with static kernel (#5775)

### What this PR does / why we need it?
This is a part of
https://github.com/vllm-project/vllm-ascend/issues/4715#issue-3694310762
1. refactor the npugraph_ex config，modified the default configuration of
the static kernel, new default value of static kernel is false
2. support online-infer with static kernel
3. fixed the issue where manually modifying FX graphs caused an abnormal
model return type, and removed the related redundant code.

### Does this PR introduce _any_ user-facing change?
yes，the new config of npugraph_ex is as follow:
```
additional_config={
            "npugraph_ex_config": {
                "enable": True,
                "enable_static_kernel": False
            }
        }
```
### How was this patch tested?
```
vllm serve /data/DeepSeek-V3.1-Terminus-w4a8 \
    --host 0.0.0.0 \
    --port 8004 \
    --data-parallel-size 4 \
    --tensor-parallel-size 4 \
    --quantization ascend \
    --seed 1024 \
    --served-model-name deepseek_v3 \
    --enable-expert-parallel \
    --max-num-seqs 48 \
    --max-model-len 40000 \
    --async-scheduling \
    --max-num-batched-tokens 9000 \
    --trust-remote-code \
    --no-enable-prefix-caching \
    --speculative-config '{"num_speculative_tokens": 3, "method":"deepseek_mtp","disable_padded_drafter_batch": false}' \
    --gpu-memory-utilization 0.9 \
    --compilation-config '{"cudagraph_capture_sizes":[4,32,64,112,160,176,192], "cudagraph_mode": "FULL_DECODE_ONLY"}' \
    --additional-config \
    '{"enable_shared_expert_dp": true,"multistream_overlap_shared_expert": true,"npugraph_ex_config":{"enable":true}}'
```

- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef

---------

Signed-off-by: chencangtao <chencangtao@huawei.com>
Signed-off-by: ChenCangtao <50493711+ChenCangtao@users.noreply.github.com>
Co-authored-by: chencangtao <chencangtao@huawei.com>

vllm_ascend/ascend_config.py

@@ -102,7 +102,8 @@ class AscendConfig:
         from vllm_ascend.utils import get_flashcomm2_config_and_validate
 
         self.flashcomm2_oproj_tensor_parallel_size = get_flashcomm2_config_and_validate(self, vllm_config)
-        self.enable_npugraph_ex = additional_config.get("enable_npugraph_ex", False)
+        npugraph_ex_config = additional_config.get("npugraph_ex_config", {})
+        self.npugraph_ex_config = NpugraphExConfig(**npugraph_ex_config)
         # We find that _npu_paged_attention still performs better than
         # npu_fused_infer_attention_score in some cases. We allow to execute
         # _npu_paged_attention in this cases. This should be removed once
@@ -211,6 +212,36 @@ class AscendFusionConfig:
         self.fusion_ops_gmmswigluquant = fusion_ops_gmmswigluquant
 
 
+class NpugraphExConfig:
+    """
+    Configuration for controlling the behavior of npugraph_ex backend.
+
+    This class provides a way to configure whether to use the npugraph_ex backend and static kernel.
+    These configurations can directly impact the performance and behavior of models deployed on Ascend platforms.
+    """
+
+    def __init__(self, enable: bool = False, enable_static_kernel: bool = False, **kwargs):
+        """
+        Initialize the configuration.
+
+        Args:
+            enable (bool): Whether to enable npugraph_ex backend.
+                When set to True, the Fx graph generated by Dymano will be
+                optimized and compiled by the npugraph_ex backend.
+                Default: False
+            enable_static_kernel (bool): Whether to enable static kernel.
+                Static kernel is suitable for scenarios with purely static shapes
+                or minimal shape changes, and can improve network performance.
+                When set to True, when during graph capture, it will compile operator
+                binary files with the corresponding shapes based on the current batch_size,
+                which usually takes some time.
+                Default: False
+            **kwargs: Additional optional parameters for forward compatibility and configuration extension.
+        """
+        self.enable = enable
+        self.enable_static_kernel = enable_static_kernel
+
+
 class XliteGraphConfig:
     """
     Configuration Object for xlite_graph_config from additional_config