vllm-ascend support Ascend950 with Qwen dense model. (#4228)

### What this PR does / why we need it?
vllm-ascend support Ascend950 with Qwen dense model
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?


- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: wangyao <iwangyao@outlook.com>
Co-authored-by: weijinqian0 <1184188277@qq.com>

vllm_ascend/worker/worker_v1.py

@@ -52,8 +52,9 @@ from vllm_ascend.device_allocator.camem import CaMemAllocator
 from vllm_ascend.distributed.parallel_state import init_ascend_model_parallel
 from vllm_ascend.ops.triton.triton_utils import init_device_properties_triton
 from vllm_ascend.platform import NPUPlatform
-from vllm_ascend.utils import (check_ascend_device_type, enable_sp,
-                               is_enable_nz, register_ascend_customop)
+from vllm_ascend.utils import (AscendDeviceType, check_ascend_device_type,
+                               enable_sp, get_ascend_device_type, is_enable_nz,
+                               register_ascend_customop)
 from vllm_ascend.worker.model_runner_v1 import NPUModelRunner
 
 torch._dynamo.trace_rules.clear_lru_cache()  # noqa: E402
@@ -87,7 +88,8 @@ class NPUWorker(WorkerBase):
         # Register ops when worker init.
         from vllm_ascend import ops
         ops.register_dummy_fusion_op()
-        _register_atb_extensions()
+        if get_ascend_device_type() != AscendDeviceType._910_95:
+            _register_atb_extensions()
         register_ascend_customop(vllm_config)
         # init ascend config and soc version
         init_ascend_config(vllm_config)
@@ -356,7 +358,8 @@ class NPUWorker(WorkerBase):
             self.model_runner.capture_model()
         # Call ATB matmul to warm up; otherwise, the first operation (ReshapeAndCache)
         # may cause performance degradation at runtime.
-        self._warm_up_atb()
+        if get_ascend_device_type() != AscendDeviceType._910_95:
+            self._warm_up_atb()
         # Reset the seed to ensure that the random state is not affected by
         # the model initialization and profiling.
         NPUPlatform.seed_everything(self.model_config.seed)