[5/N] pass the whole config to model (#9983)

Signed-off-by: youkaichao <youkaichao@gmail.com>

vllm/model_executor/models/qwen2_moe.py

@@ -30,7 +30,7 @@ from transformers import PretrainedConfig
 
 from vllm.attention import Attention, AttentionMetadata
 from vllm.compilation.decorators import support_torch_compile
-from vllm.config import CacheConfig
+from vllm.config import CacheConfig, VllmConfig
 from vllm.distributed import (get_pp_group,
                               get_tensor_model_parallel_world_size,
                               tensor_model_parallel_all_reduce)
@@ -379,11 +379,13 @@ class Qwen2MoeForCausalLM(nn.Module, SupportsPP):
 
     def __init__(
         self,
-        config: PretrainedConfig,
-        cache_config: Optional[CacheConfig] = None,
-        quant_config: Optional[QuantizationConfig] = None,
+        vllm_config: VllmConfig,
+        prefix: str = "",
     ) -> None:
         super().__init__()
+        config = vllm_config.model_config.hf_config
+        cache_config = vllm_config.cache_config
+        quant_config = vllm_config.quant_config
         self.config = config
         self.quant_config = quant_config
         self.model = Qwen2MoeModel(config, cache_config, quant_config)