Remove residual FIXME from window merge work

During window merge, DQA was temporarily disabled with GPORCA.
After the following PR, this was resolved. In this PR, I am cleaning a
residual outdated fixme

https://github.com/greenplum-db/gpdb/pull/4086

The string_agg function was categorized as ordered window aggregate in
GPDB 5. So GPORCA did not support that and fallbacks to planner. In GPDB
6 string_agg is a straightforward window aggregate, so GPORCA supports
it.

src/test/regress/expected/bfv_aggregate.out

@@ -1570,6 +1570,24 @@ select array_agg(a order by b desc nulls last) from aggordertest;
  {NULL,3,1,2,1,2}
 (1 row)
 
+-- begin MPP-14125: if prelim function is missing, do not choose hash agg.
+create temp table mpp14125 as select repeat('a', a) a, a % 10 b from generate_series(1, 100)a;
+NOTICE:  Table doesn't have 'DISTRIBUTED BY' clause -- Using column(s) named 'a' as the Greenplum Database data distribution key for this table.
+HINT:  The 'DISTRIBUTED BY' clause determines the distribution of data. Make sure column(s) chosen are the optimal data distribution key to minimize skew.
+explain select string_agg(a) from mpp14125 group by b;
+                                              QUERY PLAN                                              
+------------------------------------------------------------------------------------------------------
+ Gather Motion 2:1  (slice2; segments: 2)  (cost=8.32..9.20 rows=5 width=36)
+   ->  GroupAggregate  (cost=8.32..9.20 rows=5 width=36)
+         Group By: b
+         ->  Sort  (cost=8.32..8.57 rows=50 width=55)
+               Sort Key: b
+               ->  Redistribute Motion 2:2  (slice1; segments: 2)  (cost=0.00..5.00 rows=50 width=55)
+                     Hash Key: b
+                     ->  Seq Scan on mpp14125  (cost=0.00..3.00 rows=50 width=55)
+(8 rows)
+
+-- end MPP-14125
 -- CLEANUP
 set client_min_messages='warning';
 drop schema bfv_aggregate cascade;

src/test/regress/expected/bfv_aggregate_optimizer.out

@@ -1569,6 +1569,25 @@ select array_agg(a order by b desc nulls last) from aggordertest;
  {NULL,3,1,2,1,2}
 (1 row)
 
+-- begin MPP-14125: if prelim function is missing, do not choose hash agg.
+create temp table mpp14125 as select repeat('a', a) a, a % 10 b from generate_series(1, 100)a;
+NOTICE:  Table doesn't have 'DISTRIBUTED BY' clause. Creating a NULL policy entry.
+explain select string_agg(a) from mpp14125 group by b;
+                                                  QUERY PLAN
+--------------------------------------------------------------------------------------------------------------
+ Gather Motion 3:1  (slice2; segments: 3)  (cost=0.00..431.07 rows=10 width=8)
+   ->  Result  (cost=0.00..431.07 rows=4 width=8)
+         ->  GroupAggregate  (cost=0.00..431.07 rows=4 width=8)
+               Group Key: b
+               ->  Sort  (cost=0.00..431.06 rows=34 width=55)
+                     Sort Key: b
+                     ->  Redistribute Motion 3:3  (slice1; segments: 3)  (cost=0.00..431.01 rows=34 width=55)
+                           Hash Key: b
+                           ->  Table Scan on mpp14125  (cost=0.00..431.00 rows=34 width=55)
+ Optimizer: PQO version 2.55.0
+(10 rows)
+
+-- end MPP-14125
 -- CLEANUP
 set client_min_messages='warning';
 drop schema bfv_aggregate cascade;

src/test/regress/expected/olap_group.out

@@ -6021,22 +6021,6 @@ select sum((select prc from sale where cn = s.cn and vn = s.vn and pn = s.pn)) f
 (1 row)
 
 -- end MPP-14021
--- begin MPP-14125: if prelim function is missing, do not choose hash agg.
-create temp table mpp14125 as select repeat('a', a) a, a % 10 b from generate_series(1, 100)a;
-explain select string_agg(a) from mpp14125 group by b;
-                                              QUERY PLAN                                              
-------------------------------------------------------------------------------------------------------
- Gather Motion 2:1  (slice2; segments: 2)  (cost=8.32..9.20 rows=5 width=36)
-   ->  GroupAggregate  (cost=8.32..9.20 rows=5 width=36)
-         Group By: b
-         ->  Sort  (cost=8.32..8.57 rows=50 width=55)
-               Sort Key: b
-               ->  Redistribute Motion 2:2  (slice1; segments: 2)  (cost=0.00..5.00 rows=50 width=55)
-                     Hash Key: b
-                     ->  Seq Scan on mpp14125  (cost=0.00..3.00 rows=50 width=55)
-(8 rows)
-
--- end MPP-14125
 -- Test COUNT in a subquery
 create table prod_agg (sale integer, prod varchar);
 NOTICE:  Table doesn't have 'DISTRIBUTED BY' clause -- Using column named 'sale' as the Greenplum Database data distribution key for this table.

src/test/regress/sql/bfv_aggregate.sql

@@ -1386,7 +1386,10 @@ select array_agg(a order by b nulls last) from aggordertest;
 select array_agg(a order by b desc nulls first) from aggordertest;
 select array_agg(a order by b desc nulls last) from aggordertest;
 
-
+-- begin MPP-14125: if prelim function is missing, do not choose hash agg.
+create temp table mpp14125 as select repeat('a', a) a, a % 10 b from generate_series(1, 100)a;
+explain select string_agg(a) from mpp14125 group by b;
+-- end MPP-14125
 -- CLEANUP
 set client_min_messages='warning';
 drop schema bfv_aggregate cascade;

src/test/regress/sql/olap_group.sql

@@ -138,11 +138,6 @@ select cn, vn, pn, sum(qty*prc) from sale group by cube (cn, vn, pn) order by 1,
 select cn, vn, pn, sum(qty*prc) from sale group by grouping sets ((), (cn), (vn), (pn), (cn,vn), (cn,pn), (vn,pn), (cn,vn,pn)) order by 1,2,3; -- order 1,2,3
 --end_equiv
 
--- start_ignore
--- GPDB_84_MERGE_FIXME: ORCA started supporting this feature, investigate why
-set optimizer = off;
--- end_ignore
-
 -- ***BUG*** The extended groupings aren't correctly ordered! Maybe they wrongly parallel sorted!
 --start_equiv order 1,2,3
 select cn, vn, pn, count(distinct dt) from sale group by cn, vn, pn
@@ -177,9 +172,6 @@ order by 1,2,3; -- order 1,2,3
 select cn, vn, pn, count(distinct dt) from sale group by cube (cn, vn, pn) order by 1,2,3; -- order 1,2,3
 select cn, vn, pn, count(distinct dt) from sale group by grouping sets ((), (cn), (vn), (pn), (cn,vn), (cn,pn), (vn,pn), (cn,vn,pn)) order by 1,2,3; -- order 1,2,3
 --end_equiv
--- start_ignore
-reset optimizer;
--- end_ignore
 
 -- Ordinary Grouping Set Specifications --
 
@@ -603,18 +595,6 @@ drop table s6756 cascade; --ignore
 select sum((select prc from sale where cn = s.cn and vn = s.vn and pn = s.pn)) from sale s;
 -- end MPP-14021
 
--- begin MPP-14125: if prelim function is missing, do not choose hash agg.
-create temp table mpp14125 as select repeat('a', a) a, a % 10 b from generate_series(1, 100)a;
--- start_ignore
--- GPDB_84_MERGE_FIXME: ORCA started supporting this feature, investigate why
-set optimizer = off;
--- end_ignore
-explain select string_agg(a) from mpp14125 group by b;
--- start_ignore
-reset optimizer;
--- end_ignore
--- end MPP-14125
-
 -- Test COUNT in a subquery
 create table prod_agg (sale integer, prod varchar);
 insert into prod_agg values