[Worker][V1] Support sleep mode for v1 (#1084)

### What this PR does / why we need it?
 Support sleep mode for v1

Signed-off-by: wangli <wangli858794774@gmail.com>

vllm_ascend/platform.py

@@ -15,6 +15,7 @@
 # This file is a part of the vllm-ascend project.
 #
 
+import gc
 import logging
 import os
 from typing import TYPE_CHECKING, Optional, Tuple
@@ -118,6 +119,12 @@ class NPUPlatform(Platform):
     def mem_get_info(cls) -> Tuple[int, int]:
         return torch.npu.mem_get_info()
 
+    @classmethod
+    def clear_npu_memory(cls):
+        gc.collect()
+        torch.npu.empty_cache()
+        torch.npu.reset_peak_memory_stats()
+
     @classmethod
     def check_and_update_config(cls, vllm_config: VllmConfig) -> None:
         # initialize ascend config from vllm additional_config