Comprehensive correctness tests

[yds12]

We are just testing a few examples so far, with unit tests. This approach is not enough to guarantee that eval is 99.99% bug-free. Use something more thorough like quickcheck for that. If necessary, we can check the results of eval against other libraries such as meval.

[yds12]

We are already doing enough to close this issue:

• fuzz tests;
• tests with random valid input (using glc);
• comparison of results with other libraries (in the glc tests and in benchmarks).

[yds12]

One last kind of test we can add is one similar to the one with glc, but instead of generating valid expressions, we would generate invalid (but "almost valid") input and see if they are all rejected (with Err, never panic).