[WIP] Re-enable -dip1000 on auto-tester + fix CircleCi

[wilzbach]

Currently CircleCi fails (see e.g. #6625), because possible dlang/dmd#8408 broke the -dip1000 build of std.uni and it has been noticed, because the auto-tester didn't run the tests with -dip1000 (only CircleCi at Phobos does).
This was reverted at #6443 to unblock a now closed DMD PR.

The main problem here is that I'm still not sure how we can properly fix std.uni again.

[dlang-bot]

Thanks for your pull request, @wilzbach!

Bugzilla references

Your PR doesn't reference any Bugzilla issue.

If your PR contains non-trivial changes, please reference a Bugzilla issue or create a manual changelog.

Testing this PR locally

If you don't have a local development environment setup, you can use Digger to test this PR:

dub fetch digger
dub run digger -- build "master + phobos#6626"

[schveiguy]

I still don't think testing with DIP1000 makes sense unless we are doing the whole of Phobos.

[PetarKirov]

I still don't think testing with DIP1000 makes sense unless we are doing the whole of Phobos.

Doing the whole is intractable, so we need a way to ensure that our incremental progress doesn't regress.

[FeepingCreature]

CircleCI should be marked "not required" for the moment. A falsely failing test is worse than no test.

[wilzbach]

Okay, I will now merge this PR as CircleCi has been red for two weeks now and no one has come up with a better solution. I will open a PR for to revert the @trusted hack in std.uni, s.t. it's not forgotten and a better solution can be found.

[wilzbach]

I will open a PR for to revert the @trusted hack in std.uni, s.t. it's not forgotten and a better solution can be found.

#6638