Spine doesn't really have the concept of a pivot per se. It has a skeleton coordinate system with an origin around which the skeleton components are defined. Whether that origin coincides with the intended pivot for the skeleton is a different matter.
For SkeletonRendererComponent, which you'd use if you aren't using UMG, the location of that skeleton origin coincides with the location of the actor the component is on.
In UMG, things are different because of the way UMG works. It doesn't use the same concepts as you are used to in a "game" environment, but is geared towards UI like buttons etc. We can probably provide a function/Blueprint node that sets the alignement and anchor, but I can't think of a way to do this out of the box when you create a UMG SpineWidget. We simply do not know the sizes involved at initialization time, and once we render in the editor viewport, we have no idea if we are "allowed" to modify those properties, or if the user has already manipulated them.
Is there a reason you are using UMG as if it was a 2D game engine instead of going with actors and components?
