[feature] Illustrate EOS over-/underbinding

[eimrek]

For instance, take Hg on the website, for the AE-average, ABINIT and CASTEP: the ABINIT and CASTEP curves are always to the left of the AE average, for all 10 crystals – systematic overbinding. For Sb, there is excellent agreement for some of the 10 crystals, but for the ones where the agreement is a bit worse, ABINIT is underbinding and CASTEP is overbinding. There are other cases where the for the same element a pseudo is sometimes overbinding, something underbinding (but they seem not to be too abundant, and often it’s the XO3 that is the odd one). It could be informative to plot for a given pseudo-set a periodic table with 3 colors: systematically underbinding, systematically overbinding, mixed. That may give additional info to pseudo developers (‘is my entire set overbinding? Are some classes of elements overbinding? …)

• Suggestion by Stefaan Cottenier