Finish incomplete revert of ec63622.

The code change this made might well be fine to keep, but the
comment justifying it by reference to self-join removal isn't.
Let's just go back to the status quo ante, pending a more thorough
review/redesign of SJE.

(I found this by grepping to see if any references to self-join
removal remained in the tree.)

Author: tglsfdc

src/backend/optimizer/prep/prepunion.c

@@ -662,17 +662,14 @@ build_setop_child_paths(PlannerInfo *root, RelOptInfo *rel,
 	 * otherwise do statistical estimation.
 	 *
 	 * XXX you don't really want to know about this: we do the estimation
-	 * using the subroot->parse's original targetlist expressions, not the
+	 * using the subquery's original targetlist expressions, not the
 	 * subroot->processed_tlist which might seem more appropriate.  The reason
 	 * is that if the subquery is itself a setop, it may return a
 	 * processed_tlist containing "varno 0" Vars generated by
 	 * generate_append_tlist, and those would confuse estimate_num_groups
 	 * mightily.  We ought to get rid of the "varno 0" hack, but that requires
 	 * a redesign of the parsetree representation of setops, so that there can
-	 * be an RTE corresponding to each setop's output.  Note, we use this not
-	 * subquery's targetlist but subroot->parse's targetlist, because it was
-	 * revised by self-join removal. subquery's targetlist might contain the
-	 * references to the removed relids.
+	 * be an RTE corresponding to each setop's output.
 	 */
 	if (pNumGroups)
 	{
@@ -685,7 +682,7 @@ build_setop_child_paths(PlannerInfo *root, RelOptInfo *rel,
 			*pNumGroups = rel->cheapest_total_path->rows;
 		else
 			*pNumGroups = estimate_num_groups(subroot,
-											  get_tlist_exprs(subroot->parse->targetList, false),
+											  get_tlist_exprs(subquery->targetList, false),
 											  rel->cheapest_total_path->rows,
 											  NULL,
 											  NULL);