[MoE] [Refactor] Combine common_fused_moe and fused_moe (#3176)

### What this PR does / why we need it?
1. Move additional functionalities from fused_moe.py to
common_fused_moe.py and remove fused_moe.py
2. Remove unnecessary custom classes from qwen3_moe.py, and it will be
completely removed after we release vllm-ascend v0.11.0

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

### How was this patch tested?

Qwen3-30B-A3B/Qwen3-30B-A3B-W8A8/DeepSeek-V3-W4A8-Pruing/deepseek-mtp/pangu-pro-moe-pruing:

1. Enable/Disable EP
3. Aclgraph & eager
4. SP


- vLLM version: v0.11.0

---------

Signed-off-by: Pr0Wh1teGivee <calvin_zhu0210@outlook.com>
Co-authored-by: weijinqian0 <12153182+weijinqian0@users.noreply.github.com>

tests/ut/ops/test_fused_moe_prepare_and_finalize.py

@@ -21,6 +21,7 @@ class TestFusedMoEPrepareAndFinalize(unittest.TestCase):
         self.moe_config.tp_size = 1
         self.moe_config.ep_size = 1
         self.moe_config.dp_group = MagicMock()
+        self.moe_config.original_num_experts = 8
 
     @patch(
         "vllm_ascend.ops.moe.fused_moe_prepare_and_finalize.get_tensor_model_parallel_world_size",
@@ -196,7 +197,6 @@ class TestFusedMoEPrepareAndFinalize(unittest.TestCase):
 
         h_out, r_out, _ = layer.prepare(hidden_states,
                                         router_logits,
-                                        rm_router_logits=False,
                                         gate=mock_gate)
 
         # After all-gather with DP=2, should double the batch size
@@ -265,7 +265,6 @@ class TestFusedMoEPrepareAndFinalize(unittest.TestCase):
         # Run prepare
         h_out, r_out, _ = layer.prepare(hidden_states,
                                         router_logits,
-                                        rm_router_logits=False,
                                         gate=mock_gate)
 
         # Should be global tensor: [7, 8] and [7, 2]