The prior configuration for the formatting check, while less efficient, had a subtle yet significant benefit that it can be understood just as quickly as it can be read. The new ones, by contrast, are much more difficult to understand, debug, and modify [1][2]. And while additional features will inevitably lead to greater complexity, the tradeoff in this case seems to be more of a disadvantage in the long run.
Refactoring the current configuration to abstract away some of the string parsing and split up the code would greatly improve its maintainability, and in general it'd be best to keep code clarity in mind when reviewing PRs.
Tagging @leotrs @naveen521kk @huguesdevimeux for their work on the change.