Fix speed regression: skip no-op final boundary-slot snap in low-wave path

The final ApplyFinalBoundarySlotPolish (39s) didn't reduce violations
(4->4) but ran unconditionally. Now skipped in low-wave path.

Layout-only speed: 2m05s (down from 2m46s with optimization, was 14s
before quality pipeline). Artifact test still passes (1m50s).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

src/__Libraries/StellaOps.ElkSharp/ElkEdgeRouterIterative.WinnerRefinement.Hybrid.cs

@@ -524,7 +524,7 @@ internal static partial class ElkEdgeRouterIterative
                         continue;
                     }
 
-                    // Push above the node (smaller shift — closer to top).
+                    // Push above the node.
                     var pushY = node.Y - 1d;
                     newPath ??= path.Select(p => new ElkPoint { X = p.X, Y = p.Y }).ToList();
                     for (var pi = si; pi <= si + 1 && pi < newPath.Count; pi++)