Use non-overlapping swap for inner heapsort loop

This regresses binary-size slightly for normal builds, but the important
release_lto_thin_opt_level_s config sees a small improvement in
binary-size and a larger types such as string and 1k see 2-3% run-time
improvements with this change.

Author: Voultapher

library/core/src/slice/sort/unstable/heapsort.rs

@@ -69,9 +69,7 @@ where
                 break;
             }
 
-            // Swap `node` with the greater child, move one step down, and continue sifting. This
-            // could be ptr::swap_nonoverlapping but that adds a significant amount of binary-size.
-            ptr::swap(v_base.add(node), v_base.add(child));
+            ptr::swap_nonoverlapping(v_base.add(node), v_base.add(child), 1);
         }
 
         node = child;