Clojurians Log v2

Found a failure case pretty quickly:

(take 10 (sort-by-seed 447005435 #_(rand-int Integer/MAX_VALUE)
                         (repeatedly 1000 #(rand-int 100))))

I suspect that the key-fn can be called multiple times and it is non-deterministic (i.e., two calls with the same argument can yield different "key" values). I'm going to add logic that does the key-gen just once and produces a lookup map.

This is the fix (to ensure .nextLong is only called once per item in coll): https://github.com/seancorfield/polylith-external-test-runner/commit/db3d338b93ae523dbf3b23298981cc277754649d

No worries. It was only failing very occasionally—and always seemed to fail on the same project, but that was a red herring.