[v0.18.0][Bugfix] fix ds3.2 dcp mtp (#7681)

### What this PR does / why we need it?
Fixed the issue where the DCP overlaps the MTP scenario in the ds3.2
scenario.

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

### How was this patch tested?

cherry-pick from: https://github.com/vllm-project/vllm-ascend/pull/7617

Signed-off-by: weiguihua2 <weiguihua2@huawei.com>

tests/e2e/nightly/multi_node/config/DeepSeek-V3_2-W8A8-cp.yaml

@@ -36,7 +36,7 @@ deployment:
       --no-enable-prefix-caching
       --gpu-memory-utilization 0.85
       --trust-remote-code
-      --speculative-config '{"num_speculative_tokens": 2, "method":"deepseek_mtp"}'
+      --speculative-config '{"num_speculative_tokens": 3, "method":"deepseek_mtp"}'
       --compilation-config '{"cudagraph_capture_sizes": [3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, 36, 39, 42, 45, 48], "cudagraph_mode": "FULL_DECODE_ONLY"}' 
       --additional-config '{"layer_sharding": ["q_b_proj", "o_proj"]}'
       --tokenizer-mode deepseek_v32
@@ -62,7 +62,7 @@ deployment:
       --no-enable-prefix-caching
       --gpu-memory-utilization 0.85
       --trust-remote-code
-      --speculative-config '{"num_speculative_tokens": 2, "method":"deepseek_mtp"}'
+      --speculative-config '{"num_speculative_tokens": 3, "method":"deepseek_mtp"}'
       --compilation-config '{"cudagraph_capture_sizes": [3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, 36, 39, 42, 45, 48], "cudagraph_mode": "FULL_DECODE_ONLY"}' 
       --additional-config '{"layer_sharding": ["q_b_proj", "o_proj"]}'
       --tokenizer-mode deepseek_v32

tests/ut/attention/test_mla_v1.py

@@ -182,7 +182,7 @@ class TestAscendMLAMetadata(TestBase):
 
         metadata = AscendMLAMetadata(
             num_actual_tokens_pcp_padded, num_actual_tokens, slot_mapping,
-            query_start_loc, seq_lens, block_tables, num_decodes,
+            query_start_loc, seq_lens, seq_lens, block_tables, num_decodes,
             num_decode_tokens, num_prefills, num_input_tokens, query_lens,
             head_dim, attn_mask, attn_state, decode, prefill)
 

tests/ut/attention/test_sfa_v1.py

@@ -58,6 +58,7 @@ class TestAscendSFAMetadata(TestBase):
             num_actual_tokens=num_actual_tokens,
             slot_mapping=slot_mapping,
             seq_lens=seq_lens,
+            seq_lens_cpu=seq_lens,
             cum_query_lens=cum_query_lens,
             block_table=block_table,
             sin=sin,

tests/ut/compilation/test_acl_graph.py

@@ -803,6 +803,7 @@ class TestPCPDCPGraphParams(TestBase):
                                      slot_mapping,
                                      query_start_loc,
                                      seq_lens,
+                                     seq_lens,
                                      block_tables,
                                      4,
                                      4,