Couplings order

[oliviermattelaer]

This is based on the work of @roiser (thanks so much for it).
Trying to avoid to do what I suggest him to do (i.e. avoiding global variable).

Putting this in PR to have the CI/CD.
Next step here would be to merge my new test to check other problematic process (to see if it was the same issue)

[oliviermattelaer]

SO this

• fix the missmatch for stop pair production (confirmed by the CI )
• The status of the CI (non merged one) for e+ e- > mu+ mu- a should be consider as fixed (i.e. LTS agree with current value BUT only if setting sde_strategy=1... So they are a bug in LTS and not in the plugin for this process. So the CI might have been checking the value for sde_strategy=2, explaining the report, anyway this can be consider as fixed for this plugin.
• The CI for g g > t t~ 2g is still crashing (need investigation)

This is not as good as we would have like, but this is progress :-D
So I remove the draft status on this one.

[valassi]

Thanks @oliviermattelaer @roiser this is good to go for me.
I only checked that the susy_gg_t1t1 issue #826 is fixed, but that is confirmed.
(I attached this PR to #826 about susy and #862 about generic couplings order)
Out of the processes generated in the repo, this is the only one whose code changes.

I have opened #934 where I added code generation and I will add a few more tests.
I would prefer to merge this only after we merge also the code generation from #934.

[valassi]

This is also related to #748 (xsec mismatch in gqttq, which was also due to ther order of couplings)

[valassi]

Thanks Olivier for approving #934 :-)

I will merge this one and then merge #934.

Andrea