[OP] Enable custom op aclnnMoeInitRoutingCustom (#5332)

### What this PR does / why we need it?
This PR enables custom op `aclnnMoeInitRoutingCustom` introduced in PR
#5251

### Does this PR introduce _any_ user-facing change?
No

### How was this patch tested?

- vLLM version: release/v0.13.0
- vLLM main:
bc0a5a0c08

---------

Signed-off-by: QianChenxi <chenxi.qian.cq@outlook.com>
Signed-off-by: zzzzwwjj <1183291235@qq.com>
Co-authored-by: zzzzwwjj <1183291235@qq.com>

vllm_ascend/ops/fused_moe/token_dispatcher.py

@@ -354,7 +354,7 @@ class TokenDispatcherWithAllGather(MoETokenDispatcher):
             global_num_experts = self.num_experts_local
 
         sorted_hidden_states, expanded_row_idx, expert_tokens, pertoken_scale = (
-            torch_npu.npu_moe_init_routing_v2(
+            torch.ops._C_ascend.npu_moe_init_routing_custom(
                 hidden_states,
                 topk_ids,
                 scale=pertoken_scale,