add missed kernels

Author: William-An

src/cuda/GPU_Microbenchmark/ubench/core/lat_gmma/lat_gmma.h

@@ -264,6 +264,22 @@ void run_f32accumulator_tests() {
   run_f32tf32tf32tf32_64x128x8_test();
   run_f32tf32tf32tf32_64x192x8_test();
   run_f32tf32tf32tf32_64x256x8_test();
+  run_f32f16f16_64x8x16_test();
+  run_f32f16f16_64x16x16_test();
+  run_f32f16f16_64x32x16_test();
+  run_f32f16f16_64x64x16_test();
+  run_f32f16f16_64x96x16_test();
+  run_f32f16f16_64x128x16_test();
+  run_f32f16f16_64x192x16_test();
+  run_f32f16f16_64x256x16_test();
+  run_f32bf16bf16_64x8x16_test();
+  run_f32bf16bf16_64x16x16_test();
+  run_f32bf16bf16_64x32x16_test();
+  run_f32bf16bf16_64x64x16_test();
+  run_f32bf16bf16_64x96x16_test();
+  run_f32bf16bf16_64x128x16_test();
+  run_f32bf16bf16_64x192x16_test();
+  run_f32bf16bf16_64x256x16_test();
   run_f32e4m3e4m3e4m3_64x8x32_test();
   run_f32e4m3e4m3e4m3_64x16x32_test();
   run_f32e4m3e4m3e4m3_64x32x32_test();