Rdt energy high order correction

[oscarxblanco]

Dear all,
this PR updates the at.RDTType.CHROMATIC2 rdts linear wrt octupoles.

Expanding the energy dependence of components $b_2$ from quadrupoles and $b_3$ from sextupoles, as $(1+\delta)^{-1} = 1 - \delta + \delta^2$, sextupoles will contribute to "chromatic2" rdts with a factor $-\delta$, and quadrupoles with a factor $+\delta^2$.

Typically it is a minor correction.

Additionally, rdts dependent on dispersion and focusing terms $h11000$ and $h00110$, $h10001$, $h00002$ and $h00003$, have been added to the list of rdts.

[oscarxblanco]

Dear all, who could review this PR ?

[oscarxblanco]

As an example, I leave here below the values tested before (in red) and after (in green) the implementation of the high order correction for the hmba lattice in this repo.