Update model_loader deps and qqq quantization deps (#2220) (#2318)

Co-authored-by: HandH1998 <1335248067@qq.com>

python/sglang/srt/models/llama_embedding.py

@@ -3,10 +3,10 @@ from typing import Iterable, Tuple
 import torch
 from torch import nn
 from transformers import LlamaConfig
-from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 
 from sglang.srt.layers.pooler import EmbeddingPoolerOutput, Pooler, PoolingType
 from sglang.srt.model_executor.model_runner import ForwardBatch
+from sglang.srt.model_loader.weight_utils import default_weight_loader
 from sglang.srt.models.llama import LlamaModel
 
 
@@ -15,7 +15,6 @@ class LlamaEmbeddingModel(nn.Module):
         self,
         config: LlamaConfig,
         quant_config=None,
-        cache_config=None,
     ) -> None:
         super().__init__()
         self.model = LlamaModel(config, quant_config=quant_config)