Second iteration needed for ADTs

[iguerNL]

In the first iteration:

• extension of Alt-Ergo's native input language
• support of ADTs in the frontend (parsing, typechecking, hashconsed structure) via native language and via PSMT2
• basic theory for reasoning about ADTs

What to do next:

• review / improve / reimplement current theory
• see if we are complete, or if some reasoning steps are missing
• case-split analyse is currently disabled (and should be enabled via an option). This because it really slows down the solver
• Models generation currently not implemented for ADTs