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

python/sglang/srt/models/gemma.py

@@ -21,7 +21,7 @@ from typing import Iterable, Optional, Tuple
 import torch
 from torch import nn
 from transformers import PretrainedConfig
-from vllm.config import CacheConfig, LoRAConfig
+from vllm.config import LoRAConfig
 from vllm.distributed import get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.rotary_embedding import get_rope
 from vllm.model_executor.layers.vocab_parallel_embedding import VocabParallelEmbedding
@@ -279,7 +279,7 @@ class GemmaForCausalLM(nn.Module):
         config: PretrainedConfig,
         quant_config: Optional[QuantizationConfig] = None,
         lora_config: Optional[LoRAConfig] = None,
-        cache_config: Optional[CacheConfig] = None,
+        cache_config=None,
     ) -> None:
         del lora_config  # Unused.
         super().__init__()