[7/n] decouple quantization impl from vllm dependency - gguf kernel (#11019)

sgl-kernel/python/sgl_kernel/quantization/__init__.py

@@ -0,0 +1,8 @@
+from .gguf import (
+    ggml_dequantize,
+    ggml_moe_a8,
+    ggml_moe_a8_vec,
+    ggml_moe_get_block_size,
+    ggml_mul_mat_a8,
+    ggml_mul_mat_vec_a8,
+)