The surface temperature of fully convective stars is set by their opacity mechanism, often emission when the H minus ion is created. The rate the star emits light is not set by the contraction rate, it is the other way around-- the star loses heat at a rate set by its structure, and that's what sets the contraction rate. (As said, the rate of gravitational energy release is always double the rate the star loses heat to starlight.) The reason the surface temperature always has to be something like 3000 K is pretty complicated, but it seems to be due to the fact that this temperature is a kind of peak in the surface's ability to emit light. Convection sets the temperature structure, but there are a range of possible solutions with different internal energy, so the history of heat loss picks out whichever solution has the right internal energy. But just why the surface is always about the same T is still the trickiest part to understand, I am hoping to get a better handle on that eventually. Ignore anything you see in introductory textbooks, they never get it right.