【feat】switch for fusion ops gmmswigluquant (#5992)

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

Set a additional config parameter to control whether the gmmswigluequant
fuseion operator is enabled; it is enabled by True. / When enabled with
a small number of GPUs, the gmmswigluquant fused operator can cause some
performance degradation.

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
2c24bc6996

#### Perf

test model: GLM 4.6(w8a8)
- single A3 node(ep16, tp16),  async-scheduling, mtp, FULL_DECODE_ONLY
- bs=1, input_lens=32000, ouput_lens=1024

Without this PR: TPOT 32.22.ms
With this PR: TPOT 30.23ms

---------

Signed-off-by: zjks98 <zhangjiakang4@huawei.com>
Co-authored-by: zjks98 <zhangjiakang4@huawei.com>

vllm_ascend/ops/fused_moe/moe_comm_method.py

@@ -48,6 +48,12 @@ def setup_moe_comm_method(moe_config):
     _MoECommMethods[MoECommType.FUSED_MC2] = FusedMC2CommImpl(moe_config)
 
 
+def set_gmmswigluquant_method():
+    from vllm_ascend.ascend_config import get_ascend_config
+    ascend_config = get_ascend_config()
+    return ascend_config.ascend_fusion_config.fusion_ops_gmmswigluquant
+
+
 @dataclass
 class FusedExpertsResult:
     routed_out: torch.Tensor
@@ -69,6 +75,7 @@ class MoECommMethod(ABC):
 
         self.token_dispatcher = self._get_token_dispatcher()
         self.prepare_finalize = self._get_prepare_finalize()
+        self.use_fusion_ops = set_gmmswigluquant_method()
 
     def prepare(
         self,
@@ -159,7 +166,7 @@ class MoECommMethod(ABC):
             w2_offset=w2_offset,
             topk_scales=dispatch_results.topk_scales,
             with_quant=use_int8_w8a8 or use_int4_w4a8 or use_int4_w4a16,
-            fusion=use_int8_w8a8,
+            fusion=use_int8_w8a8 and self.use_fusion_ops,
             need_trans=need_trans,
             dynamic_eplb=dynamic_eplb)