The Doi-Edwards model in slow flows. Predictions on the Weissenberg effect