Auto merge of #112638 - lqd:rpo, r=cjgillot

Switch the BB CFG cache from postorder to RPO

The `BasicBlocks` CFG cache is interesting:
- it stores a postorder, but `traversal::postorder` doesn't use it
- `traversal::reverse_postorder` does traverse the postorder cache backwards
- we do more RPO traversals than postorder traversals (around 20x on the perf.rlo benchmarks IIRC) but it's not cached
- a couple places here and there were manually reversing the non-cached postorder traversal

This PR switches the order of the cache, and makes a bit more use of it. This is a tiny win locally, but it's also for consistency and aesthetics.

r? `@ghost`

compiler/rustc_mir_transform/src/const_prop.rs

@@ -118,8 +118,8 @@ impl<'tcx> MirPass<'tcx> for ConstProp {
 
         // Traverse the body in reverse post-order, to ensure that `FullConstProp` locals are
         // assigned before being read.
-        let postorder = body.basic_blocks.postorder().to_vec();
-        for bb in postorder.into_iter().rev() {
+        let rpo = body.basic_blocks.reverse_postorder().to_vec();
+        for bb in rpo {
             let data = &mut body.basic_blocks.as_mut_preserves_cfg()[bb];
             optimization_finder.visit_basic_block_data(bb, data);
         }

compiler/rustc_mir_transform/src/prettify.rs

@@ -24,7 +24,7 @@ impl<'tcx> MirPass<'tcx> for ReorderBasicBlocks {
 
     fn run_pass(&self, tcx: TyCtxt<'tcx>, body: &mut Body<'tcx>) {
         let rpo: IndexVec<BasicBlock, BasicBlock> =
-            body.basic_blocks.postorder().iter().copied().rev().collect();
+            body.basic_blocks.reverse_postorder().iter().copied().collect();
         if rpo.iter().is_sorted() {
             return;
         }