[Perf] fix async copy for async scheduling (#4113)

### What this PR does / why we need it?
Only CPU tensors with `pin_memory=True` can be asynchronously copied to
the device. Currently, there are two instances where non-pinned CPU
tensors are being copied to the device, which will trigger synchronous
operations, reducing the expected benefits of asynchronous scheduling.

- vLLM version: v0.11.0
- vLLM main:
83f478bb19

Signed-off-by: realliujiaxu <realliujiaxu@163.com>

vllm_ascend/worker/block_table.py

@@ -100,7 +100,7 @@ class BlockTable:
         self.slot_mapping_cpu = torch.zeros(
             self.max_num_batched_tokens +
             2 * self.pcp_world_size * self.max_num_reqs,
-            dtype=torch.int64,
+            dtype=torch.int32,
             device="cpu",
             pin_memory=self.pin_memory)
         self.slot_mapping_np = self.slot_mapping_cpu.numpy()