'True' normalisation, and some Hist2D colorbar changes

[naidoo88]

Hi Nick,

Wanted to pass along a couple of changes I've made.

Hist1D

I've added the calculation for a 'true' normalization (area under the histogram = 1), mirroring matplotlib's definition of density. So as to not interfere with any legacy code, I have called this density_norm. I have also left density as the default, but when density_norm is used, it supersedes this default behavior, with a warning message to let the user know that density has been disabled.

Further to this, I have included the uncertainty calculations for both density and density_norm. If no errorcalc is specified, but density or density_norm are used, the correct error calculation is automatically selected.

Hist2D

In order to be able to do some 'after the fact' colorbar tweaks the colorbar is now an object of the class. e.g, after plotting the Hist2D object:

my2Dhist.colorbar.ax.tick_params(which='major', width=0.7, length=2)

Further to this, to enable more control over the colorbar, I have added a 'kwargs' variable colorbar_params.
To preserve your default styling, the colorbar aspect is set to 20 when no colorbar_params argument is provided.

cbar_params={
    'pad':-0.05,
    'aspect':20
}
my2Dhist.plot(ax=ax, colorbar_params=cbar_params)