Listwise transformer unit tests are unintentionally order-independent #28
Description
The unit tests for listwise transformers are written in such a way that permutations in the order of the returned values will still pass the test
E.g. actual = [1, 2, 3], expected = [1, 3, 2] will still pass, when it should fail.
I don't think this is causing any broken functionality to pass unit tests as it stands, but we should tidy this up by changing to e.g. how tests are implemented for the new rank transformer in #26