Add a script that plots the distribution of gas particles' minimal smoothing lengths, split by z

[EvgeniiChaikin]

Here is an example of how this will look in the pipeline. The plot on the right is the new one.

[MatthieuSchaller]

The interpretation can be a bit hard I think. You could have a particles with h = 0.001 at z = 2 and then with h = 0.00001 at z=0.1. This then make it disappear from the z=1-3 panel even though it may have been the particle with the lowest h then.

[EvgeniiChaikin]

The alternative could be to not split by z at all, i.e. just create a plot that would be the same as the plot showing gas particle smoothing lengths at the snapshot's z (the left-hand figure in the description of this PR).

I am not sure which version is better. Since you added the redshift at which the minimal h is reached, I wanted to make use of it :)

[MatthieuSchaller]

It is useful but we'll have to keep that caveat in mind. An alternative is to also plot min h vs. z at which it was reached.