Remove abstime, reltime, tinterval tables from old regression databases.

In the back branches, drop these tables after the regression tests are
done with them.  This fixes failures of cross-branch pg_upgrade testing
caused by these types having been removed in v12.  We do lose the ability
to test dump/restore behavior with these types in the back branches, but
the actual loss of code coverage seems to be nil given that there's nothing
very special about these types.

Discussion: https://postgr.es/m/20181009192237.34wjp3nmw7oynmmr@alap3.anarazel.de

src/test/regress/expected/horology.out

@@ -3164,3 +3164,10 @@ SELECT to_char('2012-12-12 12:00'::timestamptz, 'YYYY-MM-DD HH:MI:SS TZ');
 (1 row)
 
 RESET TIME ZONE;
+--
+-- Drop tables that we don't want to keep because they interfere with
+-- testing pg_upgrade to v12 and up
+--
+DROP TABLE abstime_tbl;
+DROP TABLE reltime_tbl;
+DROP TABLE tinterval_tbl;

src/test/regress/expected/sanity_check.out

@@ -13,7 +13,6 @@ SELECT relname, relhasindex
    ORDER BY relname;
 a|f
 a_star|f
-abstime_tbl|f
 aggtest|f
 array_index_op_test|t
 array_op_test|f
@@ -174,7 +173,6 @@ quad_poly_tbl_ord_seq2|f
 radix_text_tbl|t
 ramp|f
 real_city|f
-reltime_tbl|f
 road|t
 shighway|t
 slow_emp4000|f
@@ -205,7 +203,6 @@ time_tbl|f
 timestamp_tbl|f
 timestamptz_tbl|f
 timetz_tbl|f
-tinterval_tbl|f
 varchar_tbl|f
 -- restore normal output mode
 \a\t

src/test/regress/sql/horology.sql

@@ -531,3 +531,12 @@ SELECT '2012-12-12 12:00 America/New_York'::timestamptz;
 SELECT to_char('2012-12-12 12:00'::timestamptz, 'YYYY-MM-DD HH:MI:SS TZ');
 
 RESET TIME ZONE;
+
+--
+-- Drop tables that we don't want to keep because they interfere with
+-- testing pg_upgrade to v12 and up
+--
+
+DROP TABLE abstime_tbl;
+DROP TABLE reltime_tbl;
+DROP TABLE tinterval_tbl;