Evaluation Task not used

[AlessandroPolidori]

Description & Motivation

I want to add this issue in order to keep in "backlog" something we have to deal with sooner or later. Evaluation Task (inside quadra/tasks/base.py) is currently extended only by SegmentationEvaluation.
I recently had the necessity to add a new behavior to the prepare() function of the evaluation tasks and it would have been much easier having a real Evaluation Task class extended by all the evaluations. I already opened the pr but is in draft rn (input_size propagation). Everything would be nicer and tidier if we refactor all the evaluation tasks and make them extend Evaluation.

[AlessandroPolidori]

I opened a pr (it's in draft because i still need to update the documentation)