Further simplify nbtree high key truncation.

Commit 7c2dbc6 reorganized _bt_truncate() in a way that enables a
further simplification that I (pgeoghegan) missed:  Since we mark the
tuple that is returned to the caller as a pivot tuple before the point
where its heap TID is set as of 7c2dbc6, it is possible to use the high
level BTreeTupleGetHeapTID() inline function to get an item pointer.  Do
it that way now.  This approach is clearer and more maintainable.

Author: petergeoghegan

src/backend/access/nbtree/nbtutils.c

@@ -2251,6 +2251,7 @@ _bt_truncate(Relation rel, IndexTuple lastleft, IndexTuple firstright,
 	tidpivot->t_info |= newsize;
 	BTreeTupleSetNAtts(tidpivot, nkeyatts);
 	BTreeTupleSetAltHeapTID(tidpivot);
+	pivotheaptid = BTreeTupleGetHeapTID(tidpivot);
 
 	/*
 	 * Lehman & Yao use lastleft as the leaf high key in all cases, but don't
@@ -2259,8 +2260,6 @@ _bt_truncate(Relation rel, IndexTuple lastleft, IndexTuple firstright,
 	 * TID.  (This is also the closest value to negative infinity that's
 	 * legally usable.)
 	 */
-	pivotheaptid = (ItemPointer) ((char *) tidpivot + newsize -
-								  sizeof(ItemPointerData));
 	ItemPointerCopy(BTreeTupleGetMaxHeapTID(lastleft), pivotheaptid);
 
 	/*