add possibility to define a complete preprocessing pipeline and execute it using Dataset.apply()

[dkrako]

Description of the problem

Currently each preprocessing step has to be executed as a single statement.

Description of a solution

It would be much nicer to add a list of preprocessing steps to a Dataset.apply() method.

Minimum acceptance criteria

• add possibility to define a complete preprocessing pipeline and execute it using Dataset.apply()