Fixes warning about configurations for rules that are not enabled, when they are enabled in a parent config

[mildm8nnered]

Fixes #4858

The problem here is configurations like this:

included:
  - Plugins
  - Source
  - Tests
excluded:
  - Tests/SwiftLintFrameworkTests/Resources
opt_in_rules:
  - implicit_return
disabled_rules:
  - type_body_length
  - function_body_length
  - large_tuple

implicit_return:
  included:
    - closure

with child configurations like this:

parent_config: .swiftlint-parent.yml

implicit_return:
  included:
    - closure
    - function
    - getter

SwiftLint will report warning: Found a configuration for 'implicit_return' rule, but it is not enabled on 'opt_in_rules'. for the child config, even though it will correctly apply the new config.

We now pass in the parent configuration when validating rule configurations for the child, and do not complain about configurations where that rule is enabled in the parent.

There should be no change in behaviour at all here - just a change in the warnings.

Getting this correct turned out to be quite complex, but after this PR, it should be the case that whenever, and only when, a rule is disabled in the parent or "child" config, a warning message will be printed, and the exact warning will vary somewhat, depending on whether the rule is disabled in the parent or "child", or not enabled at all.

There is a separate issue with child_config that this PR does not attempt to address - See #5251

[SwiftLintBot]

	17 Messages
📖	Linting Aerial with this PR took 1.22s vs 1.24s on main (1% faster)
📖	Linting Alamofire with this PR took 1.79s vs 1.81s on main (1% faster)
📖	Linting Brave with this PR took 10.36s vs 10.41s on main (0% faster)
📖	Linting DuckDuckGo with this PR took 5.52s vs 5.55s on main (0% faster)
📖	Linting Firefox with this PR took 12.78s vs 12.86s on main (0% faster)
📖	Linting Kickstarter with this PR took 12.43s vs 12.53s on main (0% faster)
📖	Linting Moya with this PR took 0.69s vs 0.69s on main (0% slower)
📖	Linting NetNewsWire with this PR took 3.66s vs 3.7s on main (1% faster)
📖	Linting Nimble with this PR took 1.01s vs 1.01s on main (0% slower)
📖	Linting PocketCasts with this PR took 10.15s vs 10.16s on main (0% faster)
📖	Linting Quick with this PR took 0.45s vs 0.46s on main (2% faster)
📖	Linting Realm with this PR took 6.36s vs 6.36s on main (0% slower)
📖	Linting Sourcery with this PR took 3.11s vs 3.13s on main (0% faster)
📖	Linting Swift with this PR took 6.21s vs 6.26s on main (0% faster)
📖	Linting VLC with this PR took 1.67s vs 1.68s on main (0% faster)
📖	Linting Wire with this PR took 23.31s vs 23.39s on main (0% faster)
📖	Linting WordPress with this PR took 15.2s vs 15.21s on main (0% faster)

Generated by 🚫 Danger

[SimplyDanny]

I lost a bit the overview. Perhaps my comments/questions are stupid. But somehow the disabled rules must be taken into account, isn't it?

[mildm8nnered]

I lost a bit the the overview. Perhaps my comments/questions are stupid. But somehow the disabled rules must be taken into account, isn't it?

I think you're right about disabled - I'll put together some local test cases to verify those, athough I'll try to find a configurable non-optin rule to test with because of #4859

[SimplyDanny]

Wow ... this has been a tough topic! After almost one year after this got opened, it will finally be merged. 😅

Thank you so much for your patience and the continuous improvements and rework!

[mildm8nnered]

Wow ... this has been a tough topic! After almost one year after this got opened, it will finally be merged. 😅

Thank you so much for your patience and the continuous improvements and rework!

No problem. We got there in the end :-)