[SYCL][CUDA] Temporarily disable flaky Regression/unoptimized_stream.cpp test on CUDA (intel/llvm-test-suite#392)

Author: yuxianch

SYCL/Regression/unoptimized_stream.cpp

@@ -3,6 +3,9 @@
 // RUN: %CPU_RUN_PLACEHOLDER %t.out
 // RUN: %GPU_RUN_PLACEHOLDER %t.out
 
+// UNSUPPORTED: cuda
+// Disable test due to flaky failure on CUDA(issue #387)
+
 // NOTE: The libclc target used by the CUDA backend used to generate atomic load
 //       variants that were unsupported by NVPTX. Even if they were not used
 //       directly, sycl::stream and other operations would keep the invalid