Parallelism for slither-mutate

[bsamuels453]

Describe the desired feature

It would be nice if there was some level of parallelism for slither-mutate. Some campaigns require multiple days if the test suite is large enough. Parallelism would be configured using a CLI arg.

[bohendo]

If tests are being run in parallel (via --parallel in hardhat or --threads in foundry) then we're likely already saturating our cores & would get little to nothing out of parallelizing the mutation side too.

A big win for performance was skipping minor mutations if major ones were uncaught (Assumes that swapping a + for a - will still pass if removing the expression entirely or replacing it with a revert passed).

We still have room for performance improvements via similar heuristics and eliminating useless mutations (eg). But mutation testing is inherently time-consuming; there's not much we can do if we want to analyze thousands of SLOC and the test suite is sluggish.. Except for renting a big beefy server and running the mutation via cloudexec 😉