Adds automated linting

[elland]

In the interest of keeping things automated, consistent and avoiding repetition, we want to automate linting.

As it can be a contentious topic, this PR will serve as both a battleground for bikeshedding the linter rules as well as the de-facto linting process, whereas we incrementally apply the newly defined rules to the whole codebase.

Checklist

• The PR Title explains the impact of the change.
• The PR description provides context as to why the change should occur and what the code contributes to that effect. This could also be a link to a JIRA ticket or a Github issue, if there is one.

[fisx]

i have a change request, but i'll approve anyway in the hope that you'll follow it anyway before merging. if you disagree i'll find out how strongly i feel about this. :)

[smatting]

I'm always surprised that there is no alternative to not . null

[elland]

Indeed. Also sounds like that's a hard naming problem. Any ideas?

[stephen-smith]

isJust . nonEmpty ? ;)

isJust . uncons ?

[elland]

How about isNotEmpty = not . null? Or to keep it consistent: notNull = not . null :D

[smatting]

+1 for keeping redundant $ and do. All the other changes look like an improvement to me.
Maybe apply hlint to more code to see more examples?

[jschaul]

Generally looks like good improvements. See in-line comments.

[jschaul]

So can we have a way to ignore enforcement of point-free style in .hlint?

Also @elland the changes here don't pass ormolu (linters step on CI); which suggests that your editor isn't set up to run ormolu on file save. You can manually run make format, but I'd suggest to hook ormolu into your editor setup. That's documented under /docs/developer somewhere.

[jschaul]

Could we get the previously discussed and approved changeset (squash-) merged, and all follow-up changes made in a separate PR, please? It's very hard to review an ever-growing PR again and again.

[elland]

@jschaul I separated them per commits, so only the latest commit needs review, but I can break it down into PRs as well.