Affected test queries have been testing the wrong thing since their
introduction in commit 4c1383efd132e4f532213c8a8cc63a455f55e344.
Back-patch to 9.3 (all supported versions).

Print columns as "column C of <relation>" rather than "<relation> column
C".  This seems to read noticeably better in English, as evidenced by the
regression test output changes, and the code change also makes it possible
for translators to adjust the phrase order in other languages.

Also change the output for OCLASS_DEFAULT from "default for %s" to
"default value for %s".  This seems to read better and is also more
consistent with the output of, for instance, getObjectTypeDescription().

Kyotaro Horiguchi, per a complaint from me

Discussion: https://postgr.es/m/20180522.182020.114074746.horiguchi.kyotaro@lab.ntt.co.jp

Make separate columns "Collation", "Nullable", "Default".

Reviewed-by: Kuntal Ghosh <kuntalghosh.2007@gmail.com>

We'll have to settle for just listing the extensions' data types,
since function arguments seem to sort differently in different locales.
Per buildfarm results.

When we implemented extensions, we made findDependentObjects() treat
EXTENSION dependency links similarly to INTERNAL links.  However, that
logic contained an implicit assumption that an object could have at most
one INTERNAL dependency, so it did not work correctly for objects having
both INTERNAL and DEPENDENCY links.  This led to failure to drop some
extension member objects when dropping the extension.  Furthermore, we'd
never actually exercised the case of recursing to an internally-referenced
(owning) object from anything other than a NORMAL dependency, and it turns
out that passing the incoming dependency's flags to the owning object is
the Wrong Thing.  This led to sometimes dropping a whole extension silently
when we should have rejected the drop command for lack of CASCADE.

Since we obviously were under-testing extension drop scenarios, add some
regression test cases.  Unfortunately, such test cases require some
extensions (duh), so we can't test for problems in the core regression
tests.  I chose to add them to the earthdistance contrib module, which is
a good test case because it has a dependency on the cube contrib module.

Back-patch to 9.1.  Arguably these are pre-existing bugs in INTERNAL
dependency handling, but since it appears that the cases can never arise
pre-9.1, I'll refrain from back-patching the logic changes further than
that.

This isn't fully tested as yet, in particular I'm not sure that the
"foo--unpackaged--1.0.sql" scripts are OK.  But it's time to get some
buildfarm cycles on it.

sepgsql is not converted to an extension, mainly because it seems to
require a very nonstandard installation process.

Dimitri Fontaine and Tom Lane

for the recent may/might cleanup.