Cons getting, setting and unification

[matko]

While there is code for getting, setting and unifying lists with prolog terms, there's nothing for the more primitive cons cells that underly these lists. This is annoying, because not every data structure that can be built with cons cells is necessarily going to be a list. Furthermore, certain approaches to list building are easier when we can leave the tail as a variable which is then unified later.

We therefore need a way to work with cons cells directly. This'll also help in implementing the cons part of #5.

See #4 for earlier discussion on this.

[ThePuzzlemaker]

You can implement this similarly to unifying with a list, just instead of unifying the end with a nil element, you can unify it with a user-specified tail. That way you're only using the "linked" part of the linked list for the part that's necessary thus not sacrificing performance by requiring use of some linked list structure.

In fact, regular lists are a special case of this list of terms + a tail unification, where the tail is nil.