[Refactor] Add expert processed token count output for DispatchFFNCombine/DispatchFFNCombineBF16 (#6402)

### What this PR does / why we need it?
Add New Output for Expert Token Count
An additional output tensor expert_token_nums is added to both operators
to meet the requirement of tracking token distribution among experts:

Tensor Name: expert_token_nums
Dimension: 1D tensor
Shape: (local_expert_num,)
Data Type: int32
Semantics: Represents the number of tokens actually received by each
expert on the current card.
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.14.1
- vLLM main:
dc917cceb8

---------

Signed-off-by: guanguan0308 <1546542263@qq.com>
Signed-off-by: guanguan0308 <162653673+guanguan0308@users.noreply.github.com>

tests/e2e/nightly/single_node/ops/multicard_ops_a3/test_dispatch_ffn_combine.py

@@ -126,6 +126,7 @@ class TestDisptachFFNCombine:
             scale2_npu.append(scale2[i].npu())
 
         out = self.generate_random_tensor((m, k), dtype=torch.bfloat16).npu()
+        expert_token_nums = self.generate_random_tensor((1, e), dtype=torch.int32).npu()
 
         torch.ops._C_ascend.dispatch_ffn_combine(
             x=x,
@@ -138,6 +139,7 @@ class TestDisptachFFNCombine:
             group=self.hcomm_info,
             max_output_size=512,
             out=out,
+            expert_token_nums=expert_token_nums,
         )
         return True
 
@@ -177,6 +179,7 @@ class TestDisptachFFNCombine:
         scale2_npu.append(scale2.npu())
 
         out = self.generate_random_tensor((m, k), dtype=torch.bfloat16).npu()
+        expert_token_nums = self.generate_random_tensor((1, e), dtype=torch.int32).npu()
 
         torch.ops._C_ascend.dispatch_ffn_combine(
             x=x,
@@ -189,6 +192,7 @@ class TestDisptachFFNCombine:
             group=self.hcomm_info,
             max_output_size=512,
             out=out,
+            expert_token_nums=expert_token_nums,
         )
         return True
 

tests/e2e/nightly/single_node/ops/multicard_ops_a3/test_dispatch_ffn_combine_bf16.py

@@ -126,6 +126,7 @@ class TestDisptachFFNCombine:
             scale2_npu.append(scale2[i].npu())
 
         out = self.generate_random_tensor((m, k), dtype=torch.bfloat16).npu()
+        expert_token_nums = self.generate_random_tensor((1, e), dtype=torch.int32).npu()
 
         torch.ops._C_ascend.dispatch_ffn_combine(
             x=x,
@@ -138,6 +139,7 @@ class TestDisptachFFNCombine:
             group=self.hcomm_info,
             max_output_size=512,
             out=out,
+            expert_token_nums=expert_token_nums,
         )
         return True
 
@@ -177,6 +179,7 @@ class TestDisptachFFNCombine:
         scale2_npu.append(scale2.npu())
 
         out = self.generate_random_tensor((m, k), dtype=torch.bfloat16).npu()
+        expert_token_nums = self.generate_random_tensor((1, e), dtype=torch.int32).npu()
 
         torch.ops._C_ascend.dispatch_ffn_combine(
             x=x,
@@ -189,6 +192,7 @@ class TestDisptachFFNCombine:
             group=self.hcomm_info,
             max_output_size=512,
             out=out,
+            expert_token_nums=expert_token_nums,
         )
         return True