myMap-infer is passing, but folds-infer isn't!

This means that we don't have to create akward equalities between
polytypes anymore

All of the modules now have appropriate export lists that give
an overview of each module.

And a lot more. We can typecheck data constructor patterns now (at
least some of them)

Everything is working perfectly! Although the code is incredibly ugly
and only supports Infer cases. I'll have to refactor it tomorrow.

Although there seems to be a very very bad bug in the solver ...
the traces show the solver is doing things like:

mv8 := mv8 -> mv8

We only have to build implications when either one of skolems or givens
is non-empty.

- Wrote pretty printing fucntions for most of the datatypes
- Added (automated) support for tracing. If you want tracing to be
  enabled, you can set the value of the traceFlag variable to True,
  and then you can find the trace of the testcase in the "traces"
  directory.
- The solver isn't halting anymore.

I might add it back again, but for now, I don't see a reason to have
it.

If a type variable is already in scope, we create a fresh type
variable, a substitution mapping the old one to a skolem that wraps
the new tyvar instead of the old one.

Added pattern synonym AppS that matches both `Arrow sg ph` and
`App sg ph`

Fixed the signature of canonEq
Got rid of topVars, and now all of the variables (including local
ones)

live in the `variables` field of the `Delta` datatype

Added more tests