Auto merge of rust-lang#107894 - Voultapher:improve-heapsort-fallback, r=scottmcm

Speedup heapsort by 1.5x by making it branchless

`slice::sort_unstable` will fall back to heapsort if it repeatedly fails to find a good pivot. By making the core child update code branchless it is much faster. On Zen3 sorting 10k `u64` and forcing the sort to pick heapsort, results in:

455us -> 278us

Author: bors

library/core/src/slice/sort.rs

@@ -198,8 +198,11 @@ where
             }
 
             // Choose the greater child.
-            if child + 1 < v.len() && is_less(&v[child], &v[child + 1]) {
-                child += 1;
+            if child + 1 < v.len() {
+                // We need a branch to be sure not to out-of-bounds index,
+                // but it's highly predictable.  The comparison, however,
+                // is better done branchless, especially for primitives.
+                child += is_less(&v[child], &v[child + 1]) as usize;
             }
 
             // Stop if the invariant holds at `node`.