[LayoutNG] Remove optimization that practically does nothing.

This was an over-otimization, which realistically probably never gets
hit.

Bug: 635619
Change-Id: Icac53b8605e56242b8455228d0767b2dafb66b5d
Reviewed-on: https://chromium-review.googlesource.com/c/1305133
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#604211}

third_party/blink/renderer/core/layout/ng/ng_block_layout_algorithm.cc

@@ -1323,18 +1323,12 @@ bool NGBlockLayoutAlgorithm::HandleInflow(
   //
   // The resulting margin strut in the above example will be {40, -30}. See
   // ComputeInflowPosition for how this end margin strut is used.
-  bool empty_block_affected_by_clearance_needs_relayout = false;
   if (empty_block_affected_by_clearance) {
     NGMarginStrut margin_strut;
     margin_strut.Append(child_data.margins.block_start,
                         child.Style().HasMarginBeforeQuirk());
 
-    // We only need to relayout if the new margin strut is different to the
-    // previous one.
-    if (child_data.margin_strut != margin_strut) {
-      child_data.margin_strut = margin_strut;
-      empty_block_affected_by_clearance_needs_relayout = true;
-    }
+    child_data.margin_strut = margin_strut;
   }
 
   // We need to layout a child if we know its BFC block offset and:
@@ -1342,8 +1336,7 @@ bool NGBlockLayoutAlgorithm::HandleInflow(
   //  - It has some unpositioned floats.
   //  - It was affected by clearance.
   if ((layout_result->Status() == NGLayoutResult::kBfcBlockOffsetResolved ||
-       relayout_child_when_bfc_resolved ||
-       empty_block_affected_by_clearance_needs_relayout) &&
+       relayout_child_when_bfc_resolved || empty_block_affected_by_clearance) &&
       child_bfc_block_offset) {
     NGConstraintSpace new_child_space = CreateConstraintSpaceForChild(
         child, child_data, child_available_size_, child_bfc_block_offset);