As in this question there were two questions, I extract the second question here.
I don't know if they spend a lot of time doing point-value memorizing, but I would imagine that there is a reasonable amount of time spent studying and memorizing endgame tesuji. Counting is important to properly evaluate when the tesuji are profitable.
If I don't mistake professionals learn a lot of typical local value of moves it's a way to play faster and correctly endgames I suppose.
Possibly the best book on endgames (for amateurs) was written by Tomoko Ogawa: http://www.gobooks.info/g15.html
She doesn't exactly "memorize" endgame positions, but she gives a list a "typical" (and not so typical) positions, and teaches us how to COUNT them.
My impression is more that the emphasis is on reading (or counting) the endgame, rather than memorizing the endgame. Certainly in that process you see some positions so frequently that you basically memorize how they are going to fall out, but this isn't quite the same thing as deliberately setting out to memorize the position.
This is especially the case because of both endgame tesuji/maek and because even fairly straightforward sequences can be interrupted due to the surrounding stones or thanks to the value of sente.