support aclgraph (#426)

<!--  Thanks for sending a pull request!

BEFORE SUBMITTING, PLEASE READ
https://docs.vllm.ai/en/latest/contributing/overview.html

-->
### What this PR does / why we need it?
<!--
- Please clarify what changes you are proposing. The purpose of this
section is to outline the changes and how this PR fixes the issue.
If possible, please consider writing useful notes for better and faster
reviews in your PR.

- Please clarify why the changes are needed. For instance, the use case
and bug description.

- Fixes #
-->
This PR supports the access of vllm-acend to the piecewise_graph feature
provided by the v1 engine.

1. register unifiled_ascend_attention_with_output for piecewise_graph to
split graph.
2. support NPUGraph to accelerate kernel launch.

### Does this PR introduce _any_ user-facing change?
<!--
Note that it means *any* user-facing change including all aspects such
as API, interface or other behavior changes.
Documentation-only updates are not considered user-facing changes.
-->
support npugraph to default， Users can disenable the npugraph feature by
configuring enforce_eager.

This has corresponding requirements for the versions of torch_npu and
CANN, and they need to support graph capture.

### How was this patch tested?
<!--
CI passed with new added/existing test.
If it was tested in a way different from regular unit tests, please
clarify how you tested step by step, ideally copy and paste-able, so
that other reviewers can test and check, and descendants can verify in
the future.
If tests were not added, please describe why they were not added and/or
why it was difficult to add.
-->
it turn to default

---------

Signed-off-by: Bug Hunter Yan <yanpq@zju.edu.cn>
Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>
Co-authored-by: Yizhou Liu <liu_yizhou@outlook.com>

vllm_ascend/worker/worker_v1.py

@@ -23,6 +23,7 @@ from typing import Dict, List, Optional
 import torch
 import torch.nn as nn
 import torch_npu
+from torch_npu.op_plugin.atb._atb_ops import _register_atb_extensions
 from vllm import envs
 from vllm.config import VllmConfig
 from vllm.distributed import (ensure_model_parallel_initialized,
@@ -65,7 +66,9 @@ class NPUWorker(WorkerBase):
         from vllm_ascend.utils import adapt_patch
         adapt_patch()
         # Register ops when worker init.
-        from vllm_ascend import ops  # noqa: F401
+        from vllm_ascend import ops
+        ops.register_dummy_fusion_op()
+        _register_atb_extensions()
 
         super().__init__(vllm_config=vllm_config,
                          local_rank=local_rank,
@@ -179,8 +182,17 @@ class NPUWorker(WorkerBase):
         self.model_runner.load_model()
 
     def compile_or_warm_up_model(self) -> None:
+        warmup_sizes = self.vllm_config.compilation_config.compile_sizes.copy()
         if not self.model_config.enforce_eager:
-            logger.warning("Graph capture is not supported on NPU.")
+            warmup_sizes = [
+                x for x in warmup_sizes if x not in
+                self.vllm_config.compilation_config.cudagraph_capture_sizes
+            ]
+        for size in sorted(warmup_sizes, reverse=True):
+            logger.info("Compile and warming up model for size %d", size)
+            self.model_runner._dummy_run(size)
+        if not self.model_config.enforce_eager:
+            self.model_runner.capture_model()
         # Reset the seed to ensure that the random state is not affected by
         # the model initialization and profiling.
         set_random_seed(self.model_config.seed)