Run gp_toolkit early to reduce testing time of it due to less logs.

gp_toolkit test tests various log related views like gp_log_system(), etc.  If
we run the test earlier, less logs are generated and thus the test runs fater.
In my test environment, the test time reduces from ~22 seconds to 6.X seconds
with this patch. Also, I check the whole test case, this change will not affect
the test coverage.

src/test/regress/greenplum_schedule

@@ -14,6 +14,13 @@
 #   to the segments, and therefore has to run in smaller groups to avoid
 #   hitting max_connections limit on segments.
 #
+
+# gp_toolkit performs a vacuum and checks that it truncated the relation. That
+# might not happen if other backends are holding transactions open, preventing
+# vacuum from removing dead tuples. And run gp_toolkit early to make some log
+# file related tests (like gp_log_system, etc) faster.
+test: gp_toolkit
+
 # enable query metrics cluster GUC
 test: instr_in_shmem_setup
 # run separately - because slot counter may influenced by other parallel queries
@@ -90,11 +97,6 @@ test: sreh
 #test: olap_window
 #test: tpch500GB
 
-# gp_toolkit performs a vacuum and checks that it truncated the relation. That
-# might not happen if other backends are holding transactions open, preventing
-# vacuum from removing dead tuples.
-test: gp_toolkit
-
 test: db_size_functions
 
 # FIXME: These tests no longer work, because they try to set