fix: remove performance cliff for regex

[Trott]

What is the purpose of this pull request?

Fixes a regex-backtracking performance cliff for some pathological cases.

What changes did you make? (Give an overview)

Replaced .* in two places with matching lists [^(] and [^|]. Wrote tests to confirm performance cliffs in current code and their elimination with this change.

[Trott]

Converted to draft because there are some weird edge cases that will have their results changed with what is currently here. For example, '(!(|))' matches the previous regex but not the new one. Should be pretty easy to fix I hope/think.

[Trott]

Converted to draft because there are some weird edge cases that will have their results changed with what is currently here. For example, '(!(|))' matches the previous regex but not the new one. Should be pretty easy to fix I hope/think.

Hmmm...actually, it looks like it's fine because all the ones that no longer match will still match GLOB_EXTENSION_SYMBOLS_RE so there's no change in behavior as far as the end user is concerned.

There's probably an opportunity to simplify/consolidate those regexes. If it is done carefully, it might be possible to do it in a way that positively impacts performance.

[mrmlnc]

LGTM.

Right now we can skip optimizations of these regular expressions, since they are used at the earliest stage of analysis patterns: managers/tasks to separate patterns by type (static/dynamic) and providers/matchers/matcher to determinate the dynamic segment of the patterns.

[mrmlnc]

Thanks again, @Trott 🎉