I can see why hapticity changes here, but not why it is necessary for the benzene ring to bend. The resonance energy of benzene is pretty large, and there's no balancing increase in stability elsewhere--unless the decrease of hapticity increases stability, which I doubt (more delocalization=more stability, not vice versa). Also, the concept of a "bent benzene" is a bit strange to me.
I'd like to know what exactly is going on in this reaction and why it happens.
@F'x mentioned that the bent benzene may just be a way of drawing the molecule to clearly show the change of overlap. If this is the case, I'd still like to know how the hapticity changed (because my initial logic behind change of hapticity is invalid if the structure doesn't actually bend).