feat: barely unusable → barely usable etc.

[hippietrail]

Issues

N/A

Description

Somewhere yesterday I read where somebody wrote "barely un-adjective", I think "barely unusable".

My guess is this is a kind of overthinking mistake made by native speakers, it's very similar to a double negative and/or to an oxymoron. As if the writer is not sure whether to use "barely" or use the "un-" prefix and end up using both. It could be nonnative speakers with a poor understanding of how "barely" works. But from the examples I found they look more typical of native speaker errors to me.

I went with the Miscellaneous LintKind as that's what the Oxymorons linter uses, which is pretty close to this one.

This linter can be a good example for anyone that wants to incorporate a dictionary check in the "pattern" side of their linter rather than the match_to_lint side. Closing with Lrc is needed. I'd like @elijah-potter to double-check that logic too please.

It also includes a new strip_prefix_ignore_ascii_case_chars() method on &[char]. This is a method I've learned is quite valuable on strings and slices generally, but the ones in the standard library are not case insensitive and would require conversion/allocation with .to_lowercase()

The dictionary is used to check that the word is an uninflected adjective both before and after removing the prefix, thus not blindly permitting unique → nique. But there could be more subtle false positives where the result of stripping the prefix is still an adjective, but not the opposite of the original. I couldn't think of any.

For now it only tackles the un- prefix, but could be extended to support im-, in-, de-, dis- - but some of those might involve more subtleties.

Running the fuzzer on this would be appreciated!

How Has This Been Tested?

cargo test

AI Disclosure

• I am a human and didn't use any AI.
• I used LLM features of my editor, but not an agent.
• I used an AI agent interactively.
• I am an agent or I got an agent to do the work autonomously.

I asked Google Search AI for tips on getting the dictionary code in the pattern side working, and again to get my final working versions more rusty/more functional style.

If Your PR Implements or Enhances a Linter

• I made up the sentences in the unit tests.
• The sentences in the unit tests were generated by an AI.
• I'm using examples from the bug report / feature request.
• I collected real-world sentences for the unit tests.

Checklist

• I have performed a self-review of my own code
• I have added tests to cover my changes
• I have considered splitting this into smaller pull requests.