[CK TILE] Support fp8/fp16 with pk_int4_t as data types for tensors A and B (#2805)

- Add support for tensor A/B in both fp16+pk_int4_t and fp8+pk_int4_t formats
- Implement A(bf8) B(i4) support in universal GEMM
- Use new implementation for i4 to fp8 conversion in Block Scale

[ROCm/composable_kernel commit: 82890192dd]

example/ck_tile/03_gemm/script/smoke_test_mem_pipeline.sh

@@ -5,7 +5,7 @@ KNAME=1
 export CK_WARMUP=0
 export CK_REPEAT=1
 
-COMMON_ARGS='-v=2 -warmup=0 -repeat=1'
+COMMON_ARGS='-v=1 -warmup=0 -repeat=1'
 
 run_tests() {
     for m in 512 1024; do
@@ -32,5 +32,8 @@ run_tests "fp16"
 run_tests "bf16"
 run_tests "fp8"
 run_tests "bf8"
+run_tests "fp16i4"
+run_tests "fp8i4"
+run_tests "bf8i4"
 
 set +x