Remove bad continuation from the codebase

[Pierre-Sassoulas]

Description

As discussed in #3512 other tools exist, and a lot of open issues are not getting fixed because those who want it are not taking the time to fix it (The first issue about it is old enough to enter primary school to learn to read).

Type of Changes

	Type
✓	🔨 Refactoring
✓	📜 Docs

Related Issue

Close #289, #638, #747, #1148, #1179, #1943, #2301, #2304, #2944, #3565

[PCManticore]

Goodbye bad-continuation 👋 🔥

[AWhetter]

If we're going down this road (and I've been leaning towards thinking we should) then I think we should consider all of the style checks together and I think we should consider whether this deserves a major version bump.

If we're saying that other tools are better at this then I think we can say the same for most of the style checks. Doing it this way would mean that we can be more clear that pylint's purpose is now as a "functional" linter as opposed to a style checker. At the very least, because we're now publicly saying that flake8/black should be run alongside pylint, I think we should get the documentation about running these tools together written up and included in the same release.

The reason I say that it maybe deserves a major version bump is because we're starting to think differently about what pylint does (style linter vs functional linter). Some people might still rely on pylint's style checking for whatever reason and doing a major version bump would mean that we can continue to provide backported updates to those users for a short time.

How do we feel about this?

[Pierre-Sassoulas]

What I understood from the discussion was that we removed only the very high effort one, which had a lot of false-positive and a lot of maintaining work to do. If a format check is low effort and have been working for years we don't remove it (even if other tools are doing formatting too).

[coveralls]

Coverage decreased (-0.1%) to 90.338% when pulling 5ddea68 on Pierre-Sassoulas:remove-bad-continuation into 716bcc4 on PyCQA:master.

[coveralls]

Coverage decreased (-0.1%) to 90.338% when pulling 5ddea68 on Pierre-Sassoulas:remove-bad-continuation into 716bcc4 on PyCQA:master.

[coveralls]

Coverage decreased (-0.1%) to 90.338% when pulling 5ddea68 on Pierre-Sassoulas:remove-bad-continuation into 716bcc4 on PyCQA:master.

[AWhetter]

That makes sense. I'm very happy to see this check go!

[revolutionisme]

Which version of pylint is this merged to? Is it already released? I am on 2.5.3 and still getting a bad-continuation error with black formatting!

[Pierre-Sassoulas]

@revolutionisme it was merged before 2.5.1 (2 may). Could you create an issue with an example of code that trigger it?

[silpheel]

@Pierre-Sassoulas: reported a new issue with an example.