Add missing args for flashinfer 0.2.10

Signed-off-by: Xin Li. <[email protected]>

Signed-off-by: XIn Li <[email protected]>
Signed-off-by: Po-Han Huang <[email protected]>

Author: xinli-sw

vllm/model_executor/layers/quantization/modelopt.py

@@ -1286,9 +1286,14 @@ def apply(
                 gemm1_weights=layer.gemm1_weights_fp4_shuffled.data,
                 gemm1_weights_scale=layer.gemm1_scales_fp4_shuffled.data.view(
                     torch.float8_e4m3fn),
+                gemm1_bias=None,
+                gemm1_alpha=None,
+                gemm1_beta=None,
+                gemm1_clamp_limit=None,
                 gemm2_weights=layer.gemm2_weights_fp4_shuffled.data,
                 gemm2_weights_scale=layer.gemm2_scales_fp4_shuffled.data.view(
                     torch.float8_e4m3fn),
+                gemm2_bias=None,
                 output1_scale_scalar=layer.g1_scale_c.data,
                 output1_scale_gate_scalar=layer.g1_alphas.data,
                 output2_scale_scalar=layer.g2_alphas.data,