That link discusses how to do it but not why. The most likely thing that occurs to me is that iOS apps consider the user a potentially hostile actor in their threat model, which is... technically a valid model, but in the context of this thread I don't that counts as a real concern.