Correct computation of scratch slot requirements.

Each SE must be assigned equal numbers of slots and slots
must be assigned in units of whole groups.

Change-Id: I8f3677237fa6f2e2d25e3e78210c5a7a0ad792f3

runtime/hsa-runtime/core/inc/scratch_cache.h

@@ -93,6 +93,7 @@ class ScratchCache {
     size_t dispatch_size;
     size_t size_per_thread;
     uint32_t lanes_per_wave;
+    uint32_t waves_per_group;
     ptrdiff_t queue_process_offset;
     bool large;
     bool retry;