Reject missing database name in pg_regress and cohorts.

Writing "pg_regress --dbname= ..." led to a crash, because
we weren't expecting there to be no database name supplied.
It doesn't seem like a great idea to run regression tests
in whatever is the user's default database; so rather than
supporting this case let's explicitly reject it.

Per report from Xing Guo.  Back-patch to all supported
branches.

Discussion: https://postgr.es/m/CACpMh+A8cRvtvtOWVAZsCM1DU81GK4DL26R83y6ugZ1osV=ifA@mail.gmail.com

src/test/regress/pg_regress.c

@@ -2257,6 +2257,17 @@ regression_main(int argc, char *argv[], init_function ifunc, test_function tfunc
 		optind++;
 	}
 
+	/*
+	 * We must have a database to run the tests in; either a default name, or
+	 * one supplied by the --dbname switch.
+	 */
+	if (!(dblist && dblist->str && dblist->str[0]))
+	{
+		fprintf(stderr, _("%s: no database name was specified\n"),
+				progname);
+		exit(2);
+	}
+
 	if (config_auth_datadir)
 	{
 #ifdef ENABLE_SSPI