Add unique_domain and other functions to pre-process time-evolution data.

[tplobo]

Accounting transient power loads, voltages, currents, etc, requires transforming the data that defines them into objects of classes that represent them as time-evolution functions.

Mathematical 1D functions essentially define a mapping of a series of values in a Domain to a series of values in an Image (x -> y). These relationships must obey certain conditions, especially to ensure compatibility with other methods/modules, such as scipy.interpolate.interp1d. This also ensures these time-evolution representations may be added, subtracted and so on.

Creation of any object that represents the time-evolution of a physical quantity must pre-process x-values and y-values to make them qualify as domain and image, ensuring the following for each case:

• new object from raw data: the domain must be composed of unique values (no x value is repeated), e.g. using a unique_domain(x,y) function.
• new object from other objects: the images of multiple objects must be expanded to match a single domain, composed by the union of all individual domains, e.g. using a match_domains(x_set,y_set)function.

[tplobo]

Closed by #3250.