- XasObject.dim1, XasObject.dim1 are no more existing
- energy is now always dimensionless

This reverts commit 9dd8869d.

There was a bug when saving ft.radius... that can be nan or numpy array. When try to save it as a numpy array then it ends up as an array of dtype 'object'.
So to fix it we now store this as a dictionnary.
And we add a try-except section when saving keys if return in the future is unsafe for hdf5.

[orangecontrib][larch] normaize call to larch: from the main thread with some callback to QApplication.processEvents

[gui][XASObjectViewer] move all plot function in the XasObjectViewer.py and for some function add a "mean", "std" display if possible

[pymca/larch] start normalization between the two process. Update display and check values for normalization vs pre_edge

For now this is just comparaison, no internal process is made to make fit the two libraries.

[orangecontrib] rework the display for ft and kweight. kWeight is displaying nothing for the time beeing.

Everything is displayed by ft througt to 'spectrum' display: the window and the ftintensity

This require the creation of some helpers to link io and XasObject, this is the goal of xas.core.io.

Enable to display a map or the spectrum
Spectrums are only displyed with energy and mu at the moment but should evolve with time.
Maybe we would like to do the same for the map ???