feat: support DeepSeek-R1-W4AFP8 model with ep-moe mode (#7762)

Signed-off-by: yangsijia.614 <yangsijia.614@bytedance.com>

python/sglang/srt/layers/quantization/__init__.py

@@ -68,6 +68,7 @@ from sglang.srt.layers.quantization.modelopt_quant import (
 )
 from sglang.srt.layers.quantization.moe_wna16 import MoeWNA16Config
 from sglang.srt.layers.quantization.qoq import QoQConfig
+from sglang.srt.layers.quantization.w4afp8 import W4AFp8Config
 from sglang.srt.layers.quantization.w8a8_fp8 import W8A8Fp8Config
 from sglang.srt.layers.quantization.w8a8_int8 import W8A8Int8Config
 
@@ -82,6 +83,7 @@ BASE_QUANTIZATION_METHODS: Dict[str, Type[QuantizationConfig]] = {
     "moe_wna16": MoeWNA16Config,
     "compressed-tensors": CompressedTensorsConfig,
     "qoq": QoQConfig,
+    "w4afp8": W4AFp8Config,
 }
 
 # VLLM-dependent quantization methods