Nice find. You may be correct that it wasn't intentional due to the lack of documentation around the change, but I do note they updated tests for the refactored function [1], and the tests clearly show different behaviour from the original implementation [2].
I'm surprised this wasn't picked up during code review. Also surprised no unit tests for the original implementation, just the new one - if you're going to refactor something, add tests for the original implementation if there aren't any already.
Who gets promoted faster: The person who spends an extra hour or two on every code review investigating the behavior of hacky wrappers 50-year-old awkward file APIs, looking for the one and 100 critical security bug? Or The person who gets more features launched?