[bugfix] Fix dummy-run and multi-node issues in MoE routing and MTP (#4947)

### What this PR does / why we need it?

- Fix a premature `return` in `moe_init_routing_quant_v2.cpp` so the
routing kernel completes correctly instead of exiting early in certain
paths.
- Switch `FusedAlltoAllCommImpl` to use the MC2-based token dispatcher
and prepare/finalize routines, aligning MoE communication with the MC2
algorithm optimized for Ascend devices.
- Add a temporary override in `MtpProposer` to map `FUSED_ALLTOALL` back
to `ALLTOALL` until the MoE communication type selection logic is fully
finalized, avoiding incorrect behavior in dummy-run flows.
- Simplify the MoE communication selection for Ascend 910-93 in
`NPUModelRunner` by removing the EP-size guard on `FUSED_ALLTOALL`,
which fixes failures in multi-node / larger-EP configurations while
keeping MC2 routing under the configured token capacity.

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

Signed-off-by: mojave2 <chenchen145@huawei.com>

vllm_ascend/spec_decode/mtp_proposer.py

@@ -734,6 +734,9 @@ class MtpProposer(Proposer):
              num_input_tokens, self.runner.with_prefill)
 
         moe_comm_type = self.runner._select_moe_comm_method(num_input_tokens)
+        # TODO: remove this after moe_comm_type selection logic is finalized
+        moe_comm_type = (MoECommType.ALLTOALL if moe_comm_type
+                         == MoECommType.FUSED_ALLTOALL else moe_comm_type)
 
         # Enable shared_expert_dp and MTP FULL graph may cause accuracy issues.
         if scheduler_output and not self.enable_shared_expert_dp: