Implement an ANSI-aware string truncation function

[meowgorithm]

Tests are also included. Also, I haven't implemented this on Buffer yet.

I abstracted out the code for detecting ANSI start and end sequences for sake of reuse, which impacts performance very slightly (thanks @kiyonlin for the benchmarking tests). We could instead manually inline that function if we wanted for a slight performance boost.

This closes #21.

[meowgorithm]

Just noticed a small correction needed on one of my comments; I’ll fix that up in the morning.

[meowgorithm]

Alright, corrected. Let me know you if you have any thoughts or questions.

[meowgorithm]

Closing as I'm noticing some odd behavior with this. I'll reopen should I sort the issue out.

[meowgorithm]

Fixed the odd behavior by adding a reset sequence after performing a truncation. Working great on my end now.

[muesli]

We should probably also move this into a separate package and provide a writer, as we do for the other operations. I can try to do it this weekend, unless one of you feels up for the task. I hear @kiyonlin has a bit of a routine going at this point 😉

[kiyonlin]

We should probably also move this into a separate package and provide a writer, as we do for the other operations. I can try to do it this weekend, unless one of you feels up for the task. I hear @kiyonlin has a bit of a routine going at this point 😉

@muesli No problem, I can move them into a separate package after this PR is done.

But the truncate package will have two extra functions StringWithTail and BytesWithTail, right?

[muesli]

But the truncate package will have two extra functions StringWithTail and BytesWithTail, right?

Correct, that sounds reasonable. Thank you @kiyonlin!

[meowgorithm]

There are still some test cases I'd like to write, but I've otherwise reworked the truncation function using Truncate in mattn/runewidth as a model.

[muesli]

Thank you @meowgorithm, @kiyonlin!

Merged and ready to be extracted into a separate package!