Scale the workgroup count down to allow correct generation for falcon with

AMD radeon cards with lower workgroup count limit

Partially fixes #1581

kompute/src/Algorithm.cpp

@@ -387,6 +387,10 @@ Algorithm::recordDispatch(const vk::CommandBuffer& commandBuffer)
 void
 Algorithm::setWorkgroup(const Workgroup& workgroup, uint32_t minSize)
 {
+    if (workgroup[0] > 65535) {
+        fprintf(stderr, "workgroup size is %d\n", workgroup[0]);
+        fflush(stderr);
+    }
 
     KP_LOG_INFO("Kompute OpAlgoCreate setting dispatch size");