Reinforced Continual Learning for Graphs