Add dp to rdts

[oscarxblanco]

This PR adds the energy offset as option to the calculation of the optics used in AT matlab computeRDT and pyat get_rdts.

[oscarxblanco]

Dear @swhite2401 and @lfarv , who could review this PR ?
This addresses issue #887 .

[swhite2401]

@oscarxblanco this looks fine to me, are the off-momentum RDTS the same as the on-momentum ones? High order terms appear no?

[oscarxblanco]

@swhite2401 , @lfarv .
Before talking about the rdt I have a doubt. Does avlinopt requires ring.is_6d False ?
I'm getting this error

>>> ring.avlinopt(refpts=at.All)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/oblanco/Documents/public/progs/pyenv/pyenvat/lib/python3.11/site-packages/at/lattice/utils.py", line 279, in wrapper
    raise AtError('{0} needs "ring.is_6d" {1}'.format(
at.lattice.utils.AtError: avlinopt needs "ring.is_6d" False

[swhite2401]

I thought I had fixed this by calling linopt6 in avlinopt, do you run the latest master version?

[swhite2401]

By the way I discussed with @carmignani on that point and the implementation you propose is missing all the feed down effects from the off-energy orbit.
It is possible to implement but would require more work.

[oscarxblanco]

I thought I had fixed this by calling linopt6 in avlinopt, do you run the latest master version?

@swhite2401 , yes, it is the latest master

>>> at.__version__
'0.6.2.dev66+gbc8b345a'

[oscarxblanco]

By the way I discussed with @carmignani on that point and the implementation you propose is missing all the feed down effects from the off-energy orbit. It is possible to implement but would require more work.

oh, I see.
I guess I should

• calculate the closed orbit in all reference points
• get the feed-down polynom B and A for those reference points in a magnet,
• extract the quad, sextupole and octupole components
• operate those components with the off-energy optics.

Am I missing something ?

[swhite2401]

By the way I discussed with @carmignani on that point and the implementation you propose is missing all the feed down effects from the off-energy orbit. It is possible to implement but would require more work.

oh, I see. I guess I should

* calculate the closed orbit in all  reference points

* get the feed-down polynom B and A for those reference points in a magnet,

* extract the quad, sextupole and octupole components

* operate those components with the off-energy optics.

Am I missing something ?

Yes I think this is about it

[swhite2401]

Concerning the avlinopt calculation I changed the linopt function but did not remove the decorator... this explains the behavior you observe, it can be easily fixed