[Enhancement][CUDA] Avoid C7508 for CUDA backend via assigning default value...

[Enhancement][CUDA] Avoid C7508 for CUDA backend via assigning default value to `minBlocksPerMultiprocesor ` (#248)

src/target/codegen_cuda.cc

@@ -98,7 +98,7 @@ void CodeGenTileLangCUDA::PrintExtraAttrs(const PrimFunc &f) {
       // return
       return;
     }
-    stream << " __launch_bounds__(" << threadIdx_ext_int->value << ")";
+    stream << " __launch_bounds__(" << threadIdx_ext_int->value << ", 1)";
   }
 }