Description of the change

We're ignoring quite a large number of HLint's warnings/suggestions, and I thought it might be a good idea to do something about that.

Here it is, one commit per line removed from .hlint.yaml (except for the first one). I've grouped the commits into sort-of themes and ordered them approximately from what I expect will be least to most controversial (bearing in mind that there are a few suggestions left that even I didn't think would lead to improvement). A handy reference list is below, so that each one can be quibbled about individually. Maybe it would be worth it to break some of these out into smaller PRs? Should each theme be its own PR? I don't know. Let's start talking.

Commits touching more than three code files are bolded.

🧠 The no-brainers

0696af28 Remove unneeded HLint ignore directives
a7fa8610 HLint fix: "Parse error"
0fd6f53f HLint fix: "Unused LANGUAGE pragma"

🤖 Boolean hijinks

50b475c0 HLint fix: "Redundant =="
b6a2aef8 HLint fix: "Redundant if"

📎 Did you mean...

c14cb8c1 HLint fix: "Use notElem"
667d083a HLint fix: "Use String"
b8784430 HLint fix: "Use isDigit"
9f2b72a1 HLint fix: "Use unless"
2dd2764c HLint fix: "Use traverse_"
6eef8922 HLint fix: "Use /="
518c3eda HLint fix: "Use $>"
cc1bd476 HLint fix: "Use =<<"
25b5042a HLint fix: "Use maybe"
9491b718 HLint fix: "Use zipWithM_"
da6cfd3a HLint fix: "Use fromMaybe"
aa0269fd HLint fix: "Use gets"
731a33f5 HLint fix: "Use unwords"

📜 To list or not to list?

fbcb5516 HLint fix: "Use &&"
82d83183 HLint fix: "Use ++"
bb6e1622 HLint fix: "Use list literal pattern"
7ff24174 HLint fix: "Use :"

🤔 Complex simplifications

c19cacab HLint fix: "Use ."
80fd0361 HLint fix: "Use Just"
54fd3dc0 HLint fix: "Hoist not"
d2edcb24 HLint fix: "Redundant fmap"

🔢 Getting it sorted

7cf1c80e HLint fix: "Use sortOn"
f279b60a HLint fix: "Avoid reverse"

💵 Dollars and sense

11eb5b21 HLint fix: "Redundant bracket"
9540bb99 HLint fix: "Redundant $"
864346c1 HLint fix: "Move brackets to avoid $"
f208a8da HLint fix: "Use <$>"
cbaf6a08 HLint fix: "Redundant do" #4090 (comment)

☝️ Pointlessness

f506048e HLint fix: "Eta reduce"
a8f3c712 HLint fix: "Use join"
93f66a42 HLint fix: "Use fmap"
a3a477b3 HLint fix: "Use const"
e3a27227 HLint fix: "Use lambda-case"

💇 Miscellaneous style choices

39d75e4d HLint fix: "Use infix"
102a4d51 HLint fix: "Use tuple-section"
74456748 HLint fix: "Use camelCase"

Finally, some words about the also-rans:

• "Use newtype instead of data"
  This one wants single-constructor data types to be declared as newtype, which maybe sounds harmless enough, but in some of the cases we have, such as

  data ClientOptions = ClientOptions
    { clientPort :: Network.PortNumber
    }

  I think it would just obscure intent.
• "Fuse foldr/map"
• "Functor law"
• "Fuse mapM/map"
  All of these are triggered by cases where, if one of the HOFs in question takes a lambda and the other a point-free expression, composition could lead to code that is less clear.
• "Avoid lambda"
• "Avoid lambda using infix"
• "Use section"
  All of these really want things like \z -> TUnknown z b to be replaced with (`TUnknown` b), which I don't think is an improvement in cases where the function wasn't named and typed with the intent of being used infix.
• "Use record patterns"
  This is the weirdest suggestion. It gets triggered by any all-wildcard constructor pattern with three or more wildcards. I think whether to use Ctor{} or Ctor _ _ _ has much more to do with semantics than the number of wildcards—specifically, whether adding a field to the constructor would be a good reason to revisit this code.
• "Reduce duplication"
  This hint might have some value but silencing it involves substantial non-mechanical refactoring, which I just didn't want to undertake.

Checklist:

• Added the change to the changelog's "Unreleased" section with a reference to this PR (e.g. "- Made a change (#0000)")
• Added myself to CONTRIBUTORS.md (if this is my first contribution)
• Linked any existing issues or proposals that this pull request should close
• Updated or added relevant documentation
• Added a test for the contribution (if applicable)