I suspect three additional factors will make it even less likely to payback:
1) The $12B cost for the wall didn't include any infrastructure support for a solar array. Chances are, if you want to stick 1x2m panels on top, then the wall itself will have to change to support that loading (panels are light, but not when the wind blows!!)
2) The cost of large scale solar arrays, at the quoted $1.50/W is not valid for a distributed array. Costs for cabling, junction boxes, interconnects etc on a linear array will be far higher than for a typical concentrated array (which is what is normally built, ie, lots of panels in a broadly square layout). And of course, the transmission losses will be significantly larger too, as the source of your power is spread out over a huge distance.
3) What are you going to do with 580MW in the middle of the southern desert? Las Vegas2 anyone?? ;-) Otherwise you've got to get that power to someone who needs it, and that isn't going to be cheap!