[P/D][Bugfix] Layerwise stacking MTP error. (#7036)

### What this PR does / why we need it?
The community has added a cleaning mechanism for the metadata after the
main model finishes running. The MTP layer should not clean the
metadata, and a new condition has been added to avoid cleaning it.

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

### How was this patch tested?
By ci

- vLLM version: v0.16.0
- vLLM main:
4034c3d32e

---------

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

vllm_ascend/worker/model_runner_v1.py

@@ -1287,6 +1287,8 @@ class NPUModelRunner(GPUModelRunner):
         has_encoder_input = self.model_config.is_encoder_decoder and num_encoder_reqs > 0
 
         # Run forward pass
+        clear_kv_metadata = self.speculative_config is None
+        if vllm_version_is("0.16.0"):
             with (
                 record_function_or_nullcontext("forward"),
                 set_ascend_forward_context(
@@ -1306,6 +1308,28 @@ class NPUModelRunner(GPUModelRunner):
                 hidden_states = self._model_forward(
                     num_tokens_padded, input_ids, positions, intermediate_tensors, inputs_embeds, **model_kwargs
                 )
+        else:
+            with (
+                record_function_or_nullcontext("forward"),
+                set_ascend_forward_context(
+                    attn_metadata,
+                    self.vllm_config,
+                    num_tokens=num_tokens_padded,
+                    num_tokens_across_dp=num_tokens_across_dp,
+                    aclgraph_runtime_mode=cudagraph_mode,
+                    batch_descriptor=batch_desc,
+                    num_actual_tokens=scheduler_output.total_num_scheduled_tokens,
+                    model_instance=self.model,
+                    max_tokens_across_pcp=0 if self.pcp_size == 1 else self.pcp_manager.max_num_tokens_across_pcp,
+                    skip_compiled=has_encoder_input,
+                ),
+                self.maybe_get_kv_connector_output(
+                    scheduler_output, clear_metadata=clear_kv_metadata
+                ) as kv_connector_output,
+            ):
+                hidden_states = self._model_forward(
+                    num_tokens_padded, input_ids, positions, intermediate_tensors, inputs_embeds, **model_kwargs
+                )
         with record_function_or_nullcontext("post process"):
             aux_hidden_states = None
             if self.use_aux_hidden_state_outputs: