Bootstrapping scores for an uncertainty estimate

[KennethEnevoldsen]

The current implementation of the evaluators only gives a singular score. This makes it hard to see the uncertainty in the scores.

A potential solution is bootstrapping on the document level and recalculating the scores.

[KennethEnevoldsen]

We have started including this within MTEB. So it is likely that future versions of SEB will include it as well.