[refactor] replace scattered business kwargs with typed request objects and explicit stage boundaries (#7024)

### What this PR does / why we need it?
Refactor `vllm_ascend/ops/fused_moe` to replace scattered MoE business
`**kwargs` with typed request objects and explicit stage boundaries.

- Prepare, dispatch, MLP, and quant stages now have clearer ownership.
- Main MoE path no longer depends on business `kwargs.get(...)` lookups.
- Comm and dispatcher interfaces are request-only on the main path.
- UTs can assert stage-level fields directly instead of inferring
behavior indirectly.

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

### How was this patch tested?
CI passed.

---------

Signed-off-by: linfeng-yuan <1102311262@qq.com>

vllm_ascend/quantization/methods/w8a8_dynamic.py

@@ -29,6 +29,7 @@ from vllm_ascend.ascend_forward_context import _EXTRA_CTX, MoECommType
 from vllm_ascend.distributed.parallel_state import get_mc2_group
 from vllm_ascend.flash_common3_context import get_flash_common3_context
 from vllm_ascend.ops.fused_moe.experts_selector import select_experts, zero_experts_compute
+from vllm_ascend.ops.fused_moe.moe_runtime_args import build_fused_experts_input
 from vllm_ascend.utils import ACL_FORMAT_FRACTAL_NZ, maybe_trans_nz
 
 from .base import AscendLinearScheme, AscendMoEScheme, QuantType
@@ -182,7 +183,9 @@ class AscendW8A8DynamicFusedMoEMethod(AscendMoEScheme):
         log2phy: torch.Tensor | None = None,
         global_redundant_expert_num: int = 0,
         pertoken_scale: Any | None = None,
-        **kwargs,
+        activation: str = "silu",
+        apply_router_weight_on_input: bool = False,
+        mc2_mask: torch.Tensor | None = None,
     ) -> torch.Tensor:
         zero_expert_num = getattr(layer, "zero_expert_num", 0)
         zero_expert_type = getattr(layer, "zero_expert_type", None)
@@ -249,19 +252,24 @@ class AscendW8A8DynamicFusedMoEMethod(AscendMoEScheme):
             w2_scale = [layer.fused_w2_scale] if fused_scale_flag else [layer.w2_weight_scale]
 
         final_hidden_states = moe_comm_method.fused_experts(
-            hidden_states=x,
-            pertoken_scale=pertoken_scale,
-            w1=w1,
-            w1_scale=w1_scale,
-            w2=w2,
-            w2_scale=w2_scale,
-            topk_weights=topk_weights,
-            topk_ids=topk_ids,
-            use_int8_w8a8=True,
-            expert_map=expert_map,
-            log2phy=log2phy,
-            dynamic_eplb=self.dynamic_eplb,
-            mc2_mask=kwargs.get("mc2_mask"),
+            fused_experts_input=build_fused_experts_input(
+                hidden_states=x,
+                topk_weights=topk_weights,
+                topk_ids=topk_ids,
+                w1=w1,
+                w2=w2,
+                quant_type=self.quant_type,
+                dynamic_eplb=self.dynamic_eplb,
+                expert_map=expert_map,
+                global_redundant_expert_num=global_redundant_expert_num,
+                mc2_mask=mc2_mask,
+                apply_router_weight_on_input=apply_router_weight_on_input,
+                log2phy=log2phy,
+                pertoken_scale=pertoken_scale,
+                activation=activation,
+                w1_scale=[layer.fused_w1_scale] if fused_scale_flag else w1_scale,
+                w2_scale=[layer.fused_w2_scale] if fused_scale_flag else w2_scale,
+            )
         )
         if zero_expert_num > 0 and zero_expert_type is not None:
             final_hidden_states += zero_expert_result