add a tuned config and insert entries in untuned config (ROCm#1243)

Co-authored-by: Xin Huang <Xin.Huang@amd.com>
Co-authored-by: Lingpeng Jin <103567126+valarLip@users.noreply.github.com>

Author: 3 people

aiter/configs/a4w4_blockscale_tuned_gemm.csv

@@ -914,3 +914,4 @@ cu_num,M,N,K,kernelId,splitK,us,kernelName,tflops,bw,errRatio
 256,60000,4096,512,54,0,165.0566,_ZN5aiter42f4gemm_bf16_per1x32Fp4_BpreShuffle_256x256E,1524.68,3077.3,0.0
 256,3000,7168,256,40,0,17.4695,_ZN5aiter41f4gemm_bf16_per1x32Fp4_BpreShuffle_96x512E,630.24,2536.39,0.0
 256,8,2112,7168,21,0,12.8435,_ZN5aiter41f4gemm_bf16_per1x32Fp4_BpreShuffle_32x128E,18.86,594.22,0.0
+256,1,2112,7168,21,0,12.3647,_ZN5aiter41f4gemm_bf16_per1x32Fp4_BpreShuffle_32x128E,2.45,612.81,0.0

aiter/configs/a4w4_blockscale_untuned_gemm.csv

@@ -182,3 +182,8 @@ M,N,K
 4096, 8192, 1024
 8192, 8192, 1024
 16384, 8192, 1024
+1, 2112, 7168
+3000, 2112, 7168
+60000, 4096, 512
+3000, 7168, 256
+8, 2112, 7168