Next runcards update

[felixhekhorn]

Closes #248

#291 is already big enough on it's own and I expect even more changes here, because we're breaking a lot of stuff so I decided for a PR on its own ...

I won't drop n_integration_cores for the moment, because we are still missing the replacement: the actual jets runner

• rerun all noteboooks
• fix unit tests

[alecandido]

I won't drop n_integration_cores for the moment, because we are still missing the replacement: the actual jets runner

I was not planning to write an "actual" jets runner in EKO ever. But if needed I can contribute the utilities for batching to Pineko.

What were your plans about it?

(I'd like, if possible, to drop even n_integration_cores before next release, to avoid another breaking one anytime soon)

[felixhekhorn]

current status:

• drop max nf configurations: they were already not used, so here the impact is mild
• drop runcards.update(): for us this is just a renaming, so again the impact should be mild, but this might effect the outside world ...
• drop integration cores: as said, I won't do here since we're lacking an alternative at the moment - this is addressed in Improve runner API #296 and subsequent PRs in pineko (cc @andreab1997 )
• evolution points in the runcard: here the impact is stronger, I'd say, especially since this is no doubt public API - we even shuffled stuff from theory to operators ($n_{f0}$)
• drop intrinsic range: this had a huge impact, since I rooted it out at the bottom

all in all, this PR is very big (look at the number of files and we're impacting eko at it's core). This will require some extended testing for which I'd like to ask some help. We also need to decide if and what to do for the transition of the files from 0.13 to 0.14, which is a breaking change.

(I also need to fix LHA, but I believe that should be not too complicated)

[felixhekhorn]

I believe the unit test failing in #291 is protected here by

eko/src/eko/io/runcards.py

Lines 216 to 218 in 7cf9e1b

	nf_default(old_th["Q0"] ** 2.0, default_atlas(ms, ks))
	if old_th["nf0"] is None
	else old_th["nf0"],

[giacomomagni]

Can we also drop this?

eko/src/eko/matchings.py

Line 82 in 7cf9e1b

def ffns(cls, nf: int, mu2: SquaredScale):

[giacomomagni]

minor comment: here all the q2_grid are now evolgrid.

[felixhekhorn]

Can we also drop this?

eko/src/eko/matchings.py

Line 82 in 7cf9e1b

def ffns(cls, nf: int, mu2: SquaredScale):

Why? this is just an alternative constructor - which is handy at times ... (and there is no max nf business around - it just sets up correctly the thresholds ...)