Ensure independence in QCheck2's monadic combinators

[jmid]

On https://discuss.ocaml.org/t/ann-bam-a-property-based-testing-with-internal-shrinking/14661/3
there was a bug report about QCheck2's monadic operators, mistakingly filed as a Discuss comment: 😏

This makes the writing of generators easier, the shrinking is internal ensuring the shrinking won’t new random values. If you use the mondic operator of QCheck2, last time I checked it was not the case. This is why to create a generator for a pair, it is recommended to use tup2 instead of monadic operators.

I believe the problem is that QCheck2 fails to use split on the underlying RNG, so that both sides of, e.g., >>= / bind can operate on the RNG state independently without RNG queries on one side affecting the RNG state of the other side.