Fix [ban]: wrong logic matching deeply nested methods

[lpaladin]

PR checklist

• Addresses an existing issue: #0000
• New feature, bugfix, or enhancement
  • Includes tests
• Documentation update

Overview of change:

Fix a bug where a.b.c() may also match with a.c().

Is there anything you'd like reviewers to focus on?

CHANGELOG.md entry:

[bugfix] ban: Fix a false positive which would occur when banning method calls nested inside objects. Previously, banning ["a", "b", "c"] would trigger lint failures on the syntax b.c(), which was not the intent of this rule.

[palantirtech]

Thanks for your interest in palantir/tslint, @zhouhaoyu! Before we can accept your pull request, you need to sign our contributor license agreement - just visit https://cla.palantir.com/ and follow the instructions. Once you sign, I'll automatically update this pull request.

[JoshuaKGoldberg]

LGTM, thanks @zhouhaoyu!

[JoshuaKGoldberg]

Note: this would be a breaking change, as some users may be accidentally relying on the old (technically incorrect) behavior. This PR will have to wait until TSLint is ready to merge in PRs for a next major version.

Generally it'd be better to have an issue filed before a PR for changes, but this seems small enough to discuss here. Great catch @zhouhaoyu!

[alexluecke]

Any idea when this will be released?

[JoshuaKGoldberg]

/cc @adidahiya

[adidahiya]

We usually treat fixes to rule behavior like this as non-breaking, even though users might be relying on the old faulty behavior. I just want to confirm that the false positive existed before this change and then I'll merge this.

[adidahiya]

updated the changelog entry to be more precise, let me know if that looks ok