Adopt inductor fusion and define quantization fusion pass (#4168)

### What this PR does / why we need it?
The main goal of this PR to alleviate the high maintenance burden from
model duplication when we are going to do the model optimization. Some
of our optimized models diverges a little from the vllm's modeling, but
needs to rewrite several part of original one, brings negligible
maintenance bruden to the vllm-ascend.In order to solve that, we propose
to leverage `torch.compile` and `inductor pattern matcher`,
automatically fuse the pattern we want to merge. For more details can
refer to the RFC https://github.com/vllm-project/vllm-ascend/issues/4239

This pr integrates `AddRMSNorm` and the `Quant` operator, which can
improve the inference speed of models using `w8a8 `quantization.

### Does this PR introduce _any_ user-facing change?
Yes, add new additional_config

### How was this patch tested?
```python
def main():
    prompts = [
        "The president of the United States is Mr.",
    ]

    # Create a sampling params object.
    sampling_params = SamplingParams(max_tokens=100, temperature=0.6, top_k=40, top_p=0.95)
    # Create an LLM.
    llm = LLM(
        model="/root/.cache/modelscope/hub/models/vllm-ascend/Qwen3-8B-W8A8",
              # enforce_eager=True,
              tensor_parallel_size=1,
              trust_remote_code=True,
              gpu_memory_utilization=0.7,
              quantization="ascend",
              )

    # Generate texts from the prompts.
    outputs = llm.generate(prompts, sampling_params)
    for output in outputs:
        prompt = output.prompt
        generated_text = output.outputs[0].text
        print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
```

```text
Prompt: 'The president of the United States is Mr.', Generated text: ' Trump. The president of the United States is Mr. Biden. Which of the following statements is correct? \n\nA. Mr. Trump is Mr. Biden.  \nB. Mr. Trump is not Mr. Biden.  \nC. The president of the United States is not Mr. Trump.  \nD. The president of the United States is not Mr. Biden.\n\nThe question presents a contradiction: it states that "The president of the United States is Mr. Trump" and "The president of'
```


- vLLM version: 86e178f7c4d8c3b0eaf3c8e3f810a83f63b90e24
- vLLM main:
86e178f7c4

---------

Signed-off-by: Icey <1790571317@qq.com>
Signed-off-by: wxsIcey <1790571317@qq.com>

vllm_ascend/ascend_config.py

@@ -36,9 +36,15 @@ class AscendConfig:
         additional_config = vllm_config.additional_config if vllm_config.additional_config is not None else {}
         torchair_graph_config = additional_config.get("torchair_graph_config",
                                                       {})
+
         self.torchair_graph_config = TorchairGraphConfig(
             torchair_graph_config, vllm_config, additional_config)
 
+        ascend_compilation_config = additional_config.get(
+            "ascend_compilation_config", {})
+        self.ascend_compilation_config = AscendCompilationConfig(
+            **ascend_compilation_config)
+
         ascend_scheduler_config = additional_config.get(
             "ascend_scheduler_config", {})
         self.ascend_scheduler_config = AscendSchedulerConfig(
@@ -144,6 +150,31 @@ class AscendConfig:
             self, vllm_config)
 
 
+class AscendCompilationConfig:
+    """
+    Configuration for controlling the behavior of Ascend graph optimization.
+
+    This class provides a way to configure graph fusion optimizations.
+    These configurations directly impact the performance and behavior of models
+    deployed on Ascend platforms.
+    """
+
+    def __init__(self, enable_quantization_fusion: bool = True, **kwargs):
+        """
+        Initialize the configuration.
+        
+        Args:
+            enable_quantization_fusion (bool): Whether to enable quantization fusion optimization.
+                When set to True, the system will optimize quantization-related operations,
+                reducing the number of quantization/dequantization nodes.
+                Default: True
+                
+            **kwargs: Additional optional parameters for forward compatibility and configuration extension.
+        """
+        self.enable_quantization_fusion = enable_quantization_fusion
+        # Add more compilation related configs here as needed
+
+
 class TorchairGraphConfig:
     """
     Configuration Object for torchair_graph_config from additional_config
@@ -326,6 +357,11 @@ def check_ascend_config(vllm_config, enforce_eager):
                     "it has been disabled automatically.")
         # aclgraph case
         else:
+            if ascend_config.ascend_compilation_config.enable_quantization_fusion:
+                logger.info(
+                    "Quantization fusion enabled! op fusion on quantization are expected. "
+                )
+
             if vllm_config.model_config:
                 model_type = vllm_config.model_config.hf_config.model_type
                 if "qwen" not in model_type: