Additionally, refactor some code to give everything its own space.
Mainly, put any code that renames/substitutes variables in expressions
into its own module (WLP.RenameFunctions).

Note that not ALL programs in the mandatory part are done yet! We still
need Array assignment, mostly.

Some functions seem to assume that the program paths are formed by a
fully left-associative application of the Seq constructor (as in: Seq
Skip $ Seq Skip $ ... $ Seq Skip Skip). Therefore, this function is
added; it will turn any string of Seq'ed statements into this shape.
Using this function will do no harm (other than possibly spending effort
into doing nothing), so use it accordingly.

And merged it with the edited version we had. It seems that this extra
line is the only thing that changed.

Some examples had syntax errors; those have been fixed. Additionally,
I added a simple example with a while loop. It does not contain a proper
post-condition either, which might be interesting as well (as the
program might be able to handle those properly too).

This change allows us to calculate program paths from all programs (as
long as they don't contain any try-catch). Using this, we should be able
to calculate ALL wlps for the programs. However, the checks are not
complete for the whole program, as only program paths up to a certain
length are checked. Any further program paths might say that the program
is (in)correct and give a different result.

The WLP of a subset of statements in the language can now be calculated.
This needs to be extended, but it gives a basic idea of what we can do.

I made new types for the logical expressions (and the mathematical
expression in them) rather than reusing the datatype within the
language. I feel that there's a benefit to separating these, showing
when we're talking about the programs and when we're talking about the
wlp and its calculation and the like. However, this does increase the
amount of types we use. I think the benefit outweighs the cost of having
extra types, but I also feel like we should discuss this :)

Extended the GCL Parser/Lexer to also parse/lex RepBy

See merge request !1

In case anyone wants to use a sandbox, this should make sure that is not
being pushed. Also makes sure the build is not pushed.

Comes with a .cabal file that says to use z3, but you gotta install it
first. This is *kind of* involved and so we should just do that
together.