Yes, in many many posts, I said let's exclude collisions.
Also, if there is no friction there is no terminal velocity unless it is zero.
Further, as I said in a previous post. You have to be very careful if you are going to use the conservation of angular momentum argument. It is only valid around a fixed point of rotation.
Yes, you have to ski along a curved surface. Across a halfpipe will do.
Another example, take the half-pipe example where you implied you can only speed up.
Every time you come to a stop, jump to the bottom of the half pipe and absorb the perpendicular component.
Will the speed decrease or increase after each jump?
The first jump is simple and it is easy to show that at the bottom you will have a kinetic energy which is about 1/4 of what you would have if you only slided down the surface.
This mean you will only make it 1/4 up the other side. Now jump again to the bottom.
You will get an energy time series which converges to zero.
I never said that or meant to imply that. CTkook and I had many conversations about slowing down in the halfpipe and also stopping on top of the bump, all of which I agreed with. The only point I've made is that Jonny Moseley and the Canadian freestyle team say to actively pull up the legs so that your COM doesn't get pushed up. Effects that slow you down by raising your COM in many circumstances are counter productive because the dominant effect for speed control is not allowing gravity to accelerate you on steep surfaces.
True, I would maybe oversimplify this and lump it in with collisions. There are many many effects. My point is that friction is dominant enough that we can understand A&E over the broadest circumstances just by analyzing the effects of gravity and friction.