New plots for generation and other changes to plots

In order to better use the generator output, I have implemented a plot to show the performance of the new generated designs, with respect to the requested values.

First thing I have to change is that now the generator.generate not only returns the data, but the full dictionary with the information.

Then, we can leverage that in the plotter, and provide a list of outputs from the generate method, in order to obtain the following plot

As we can see, both specific requests and ranges are visualized.

Additionally, now all plots accept the flag flag_unnorm. Even though in some cases it is not too valuable, it is for the plot distrib_attributes, as now we can for example plot the outputML in the original range with flag_unnorm=True, which we could not before.

Also, in some plots I have allowed requesting subattributes, i.e. specific dimensions of data objects, such as for example the countour_2d. In this case, it is really heavy to plot all attributes, and for that reason the best can be to plot just some subattributes

or the distribution of all the dimensions of that data object

Additionally, the name of the force_update has been changed to flag_update_cat. This flag is still required, as explained in #121 (closed)