Delete segspace_setup and segspace_cleanup test

segspace test passes without need to set the GUC
gp_workfile_limit_per_segment. Just to set and reset this GUC setup
and cleanup test used to restart the cluster. Since it's not needed
delete these tests, which helps to save time. I don't have context on
why setup was needed earlier but definitely doesn't seem to be
required now.

Discussion:
https://groups.google.com/a/greenplum.org/d/msg/gpdb-dev/9lB2ovwKP3I/W6iQoN4GCAAJ

src/test/regress/greenplum_schedule

@@ -79,11 +79,8 @@ test: zlib
 test: instr_in_shmem_verify
 
 test: workfile_limits
-# This test will change the gp_workfile_limit_per_segment and will need to restart the DB.
 # It will also use faultinjector - so it needs to be in a group by itself.
-test: segspace_setup
 test: segspace
-test: segspace_cleanup
 
 test: cursor
 

src/test/regress/sql/segspace_cleanup.sql

---
--- Clean up for segspace.sql tests
--- 
-
--- start_ignore
-\! gpconfig -r gp_workfile_limit_per_segment
-\! PGDATESTYLE="" gpstop -rai
--- end_ignore

src/test/regress/sql/segspace_setup.sql

---
--- Setup for segspace.sql tests
--- 
-
--- We are setting gp_workfile_limit_per_segment to be 5.2GB so that all tests in 
--- segspace.sql can run without hitting workfile limit. The tests examine if 
--- 'used_segspace' counter is changed and get reset to 0 at the end of 
--- query execution. If gp_workfile_limit_per_segment == 0 (default value), 
--- then we don't change 'used_segspace' counter.
-
--- start_ignore 
-\! gpconfig -c gp_workfile_limit_per_segment -v 5242880
-\! PGDATESTYLE="" gpstop -rai
--- end_ignore