[model:performance] Experiment with including bugs with their performance impact is `none` in the training set only

Yeah, let's do that for now, if it is "none" it should not be considered as a perf bug.

We can file a follow-up issue to investigate using these as performance or not as performance in the training set but excluding them from the test set to see if they have any impact.

Originally posted by @marco-c in https://github.com/mozilla/bugbug/pull/3895#discussion_r1417400177

Depends on: https://github.com/mozilla/bugbug/issues/3876

See also: https://github.com/mozilla/bugbug/issues/3923