The current face index can always be retrieved from the "loop_to_face" map, so passing it around and updating it manually just adds more state to track.
The current face index can always be retrieved from the "loop_to_face" map, so passing it around and updating it manually just adds more state to track.