Evaluating the usefulness of functional distance measures when calibrating journey-to-crime distance decay functions