Handle 2^16 slabs

This code assumes that 16bit uint wrapping can be used, rather than
masking by SLAB_SIZE-1.  Add explicit mask, so we can change the
SLAB_SIZE to be smaller.

src/mem/slab.h

@@ -58,7 +58,7 @@ namespace snmalloc
       {
         // This slab is being bump allocated.
         p = (void*)((size_t)this + head - 1);
-        meta->head = head + (uint16_t)rsize;
+        meta->head = (head + (uint16_t)rsize) & (SLAB_SIZE - 1);
         if (meta->head == 1)
         {
           meta->set_full();