Don't hardcode contrib_regression dbname in postgres_fdw and dblink tests.

 Don't hardcode contrib_regression dbname in postgres_fdw and dblink tests.
 
 That allows parallel installcheck to succeed inside contrib/. The
 output is not particularly pretty unless make's -O option to
 synchronize the output is used.
 
 There's other tests, outside contrib, that use a hardcoded,
 non-unique, database name. Those prohibit paralell installcheck to be
 used across more directories; but that's something for a separate
 patch.