[1/N] Remove CacheConfig import in all model files (#1658)

python/sglang/srt/models/xverse.py

@@ -22,7 +22,6 @@ from typing import Any, Dict, Iterable, Optional, Tuple
 import torch
 from torch import nn
 from transformers import LlamaConfig
-from vllm.config import CacheConfig
 from vllm.distributed import get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.activation import SiluAndMul
 from vllm.model_executor.layers.layernorm import RMSNorm
@@ -297,7 +296,7 @@ class XverseForCausalLM(nn.Module):
         self,
         config: LlamaConfig,
         quant_config: Optional[QuantizationConfig] = None,
-        cache_config: Optional[CacheConfig] = None,
+        cache_config=None,
         efficient_weight_load=False,
     ) -> None:
         super().__init__()