Fix building tests in Xcode

[ryanaveo]

Fix #4606

SwiftLintTestHelpers is a testTarget in order to use @testable imports for SwiftLintFramework, which had previously failed to compile for release when the target was declared as a regular target with the following error:

> swift build -c release

error: module 'SwiftLintFramework' was not compiled for testing
@testable import SwiftLintFramework

However, this breaks building/testing the target in Xcode with the following error:

Unable to resolve build file: XCBCore.BuildFile (The workspace has a reference to a missing target with GUID 'PACKAGE-TARGET:SwiftLintTestHelpers')

In order for SwiftLintTestHelpers to use internal symbols in SwiftLintFramework without using @testable import, we use @_spi imports instead. This fixes the issues when building on Xcode and for release.

Test Plan

• swift build -c release

Compiling plugin Generate Manual...
Compiling plugin SwiftLintPlugin...
Building for production...
[32/32] Compiling SwiftLintTestHelpers RuleDescription+Examples.swift
Build complete! (225.61s)

• Xcode builds successfully

Notes
Rule.swift needed to be broken into two files to pass the file length rule

[SwiftLintBot]

	1 Warning
⚠️	If this is a user-facing change, please include a CHANGELOG entry to credit yourself! You can find it at CHANGELOG.md.

	18 Messages
📖	Linting Aerial with this PR took 1.2s vs 1.2s on main (0% slower)
📖	Linting Alamofire with this PR took 1.56s vs 1.58s on main (1% faster)
📖	Linting Brave with this PR took 8.21s vs 8.22s on main (0% faster)
📖	Linting DuckDuckGo with this PR took 3.16s vs 3.18s on main (0% faster)
📖	Linting Firefox with this PR took 10.15s vs 10.18s on main (0% faster)
📖	Linting Kickstarter with this PR took 11.21s vs 11.2s on main (0% slower)
📖	Linting Moya with this PR took 0.63s vs 0.64s on main (1% faster)
📖	Linting NetNewsWire with this PR took 3.53s vs 3.53s on main (0% slower)
📖	Linting Nimble with this PR took 0.69s vs 0.69s on main (0% slower)
📖	Linting PocketCasts with this PR took 8.22s vs 8.22s on main (0% slower)
📖	Linting Quick with this PR took 0.25s vs 0.26s on main (3% faster)
📖	Linting Realm with this PR took 12.52s vs 12.55s on main (0% faster)
📖	Linting SourceKitten with this PR took 0.49s vs 0.49s on main (0% slower)
📖	Linting Sourcery with this PR took 2.59s vs 2.57s on main (0% slower)
📖	Linting Swift with this PR took 4.94s vs 4.92s on main (0% slower)
📖	Linting VLC with this PR took 1.42s vs 1.43s on main (0% faster)
📖	Linting Wire with this PR took 9.68s vs 9.63s on main (0% slower)
📖	Linting WordPress with this PR took 12.18s vs 12.21s on main (0% faster)

Here's an example of your CHANGELOG entry:

* Fix building tests in Xcode.  
  [ryanaveo](https://github.com/ryanaveo)
  [#issue_number](https://github.com/realm/SwiftLint/issues/issue_number)

note: There are two invisible spaces after the entry's text.

Generated by 🚫 Danger

[jpsim]

Thanks @ryanaveo !