[New model] Qwen3-next support (#2917)

### What this PR does / why we need it?
Add Qwen3-next support.

### Does this PR introduce _any_ user-facing change?
Yes, users can use Qwen3 next.
Related doc: https://github.com/vllm-project/vllm-ascend/pull/2916 the
tutorial will be ready in
[here](https://vllm-ascend.readthedocs.io/en/latest/tutorials/multi_npu_qwen3_next.html)

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

Related: https://github.com/vllm-project/vllm-ascend/issues/2884

Co-Authored-By: Angazenn <supperccell@163.com>
Co-Authored-By: zzzzwwjj <1183291235@qq.com>
Co-Authored-By: MengqingCao <cmq0113@163.com>
Co-Authored-By: linfeng-yuan <1102311262@qq.com>
Co-Authored-By: hust17yixuan <303660421@qq.com>
Co-Authored-By: SunnyLee219 <3294305115@qq.com>
Co-Authored-By: maoxx241 <maoxx241@umn.edu>


- vLLM version: v0.10.2
- vLLM main:
b834b4cbf1

---------

Signed-off-by: MengqingCao <cmq0113@163.com>
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
Signed-off-by: Angazenn <supperccell@163.com>
Signed-off-by: Your Name <you@example.com>
Signed-off-by: zzzzwwjj <1183291235@qq.com>
Signed-off-by: linfeng-yuan <1102311262@qq.com>
Signed-off-by: hust17yixuan <303660421@qq.com>
Co-authored-by: MengqingCao <cmq0113@163.com>
Co-authored-by: Angazenn <supperccell@163.com>
Co-authored-by: Your Name <you@example.com>
Co-authored-by: zzzzwwjj <1183291235@qq.com>
Co-authored-by: linfeng-yuan <1102311262@qq.com>
Co-authored-by: hust17yixuan <303660421@qq.com>

tests/ut/attention/test_attention_v1.py

@@ -72,7 +72,8 @@ class TestAscendAttentionMetadataBuilder(TestBase):
         self.mock_vllm_config.model_config.max_model_len = 640
         self.mock_vllm_config.cache_config.block_size = 64
         self.mock_device = 'cpu:0'
-        self.builder = AscendAttentionMetadataBuilder(self.mock_vllm_config,
+        self.builder = AscendAttentionMetadataBuilder(None, None,
+                                                      self.mock_vllm_config,
                                                       self.mock_device)
 
     def test_reorder_batch(self):
@@ -105,14 +106,16 @@ class TestAscendAttentionMetadataBuilder(TestBase):
             positions=torch.tensor([10, 10]),
             attn_mask=torch.ones((10, 10)),
             spec_attn_mask=None,
-            attn_state=AscendAttentionState.PrefillNoCache)
+            attn_state=AscendAttentionState.PrefillNoCache,
+            num_computed_tokens_cpu=None,
+            seq_lens=None)
 
         mock_nz_tensor = MagicMock()
         mock_model = MagicMock()
         mock_nd_to_nz_2d.return_value = mock_nz_tensor
         mock_npu_format_cast.return_value = mock_nz_tensor
 
-        self.builder.build(common_attn_metadata, mock_model)
+        self.builder.build(1, common_attn_metadata, mock_model)
 
     @patch('vllm_ascend.attention.attention_v1.AscendMetadata')
     @patch('torch_npu.npu_format_cast')
@@ -136,7 +139,9 @@ class TestAscendAttentionMetadataBuilder(TestBase):
             positions=torch.tensor([10, 10]),
             attn_mask=torch.ones((15, 15)),
             spec_attn_mask=None,
-            attn_state=AscendAttentionState.ChunkedPrefill)
+            attn_state=AscendAttentionState.ChunkedPrefill,
+            num_computed_tokens_cpu=None,
+            seq_lens=None)
 
         mock_ascend_attention_state = MagicMock()
         mock_ascend_attention_state.PrefillNoCache = 0
@@ -146,7 +151,7 @@ class TestAscendAttentionMetadataBuilder(TestBase):
         mock_nd_to_nz_spec.return_value = mock_nz_tensor
         mock_npu_format_cast.return_value = mock_nz_tensor
 
-        self.builder.build(common_attn_metadata, mock_model)
+        self.builder.build(1, common_attn_metadata, mock_model)
 
     @patch('vllm_ascend.attention.attention_v1.AscendMetadata')
     @patch('vllm_ascend.attention.attention_v1.is_310p', return_value=False)
@@ -165,10 +170,12 @@ class TestAscendAttentionMetadataBuilder(TestBase):
             positions=torch.tensor([10, 10]),
             attn_mask=torch.ones((15, 15)),
             spec_attn_mask=None,
-            attn_state=AscendAttentionState.ChunkedPrefill)
+            attn_state=AscendAttentionState.ChunkedPrefill,
+            num_computed_tokens_cpu=None,
+            seq_lens=None)
         mock_model = MagicMock()
 
-        self.builder.build(common_attn_metadata, mock_model)
+        self.builder.build(1, common_attn_metadata, mock_model)
 
 
 class TestAscendAttentionBackendImpl(TestBase):

tests/ut/attention/test_mla_v1.py

@@ -189,7 +189,8 @@ class TestAscendMLAMetadataBuilder(TestBase):
         ascend_config = MagicMock()
         with patch("vllm_ascend.attention.mla_v1.get_ascend_config",
                    return_value=ascend_config):
-            builder = AscendMLAMetadataBuilder(mock_vllm_config, mock_device)
+            builder = AscendMLAMetadataBuilder(None, None, mock_vllm_config,
+                                               mock_device)
 
             self.assertEqual(builder.block_size,
                              mock_vllm_config.cache_config.block_size)
@@ -209,7 +210,8 @@ class TestAscendMLAMetadataBuilder(TestBase):
 
         with patch("vllm_ascend.attention.mla_v1.get_ascend_config",
                    return_value=ascend_config):
-            builder = AscendMLAMetadataBuilder(mock_vllm_config, mock_device)
+            builder = AscendMLAMetadataBuilder(None, None, mock_vllm_config,
+                                               mock_device)
             builder.decode_threshold = 1
 
         input_batch = MagicMock()