fix deepseek mla

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>

Author: wangxiyuan

vllm_ascend/models/layers/mla.py

@@ -133,7 +133,7 @@ def forward(
                              dtype=hidden_states.dtype,
                              device=hidden_states.device)
         output = self.mla_attn.impl.forward(hidden_states, kv_cache,
-                                            forward_context.attn_metadata,
+                                            forward_context.attn_metadata[self.mla_attn.layer_name],
                                             need_gather_q_kv, output)
         output = output.view(-1, output_shape[-1])
         return output