Compatibility with pineappl >= 1.0

[giacomomagni]

This PR aims to update the integrability and positivity scripts to be compatible with pineappl v 1.0.

[felixhekhorn]

Let's try to merge this! but there are still issues:

1. somehow pylint doesn't seem to understand pineappl, see e.g. here . As a temporary fix I simply deactivate it, but this is surely not the right solution. I can reproduce that error locally, but I can also do poetry run python -c "import pineappl; help(pineappl.grid.Grid.read)" and get something sane back. What is the problem?
2. vrap needs to be adjusted as well, see here
3. there are some cyclic imports going round, which might be an issue - see here

help would be appreciated @giacomomagni @Radonirinaunimi 😇

EDIT: PS: the local pylint command even generates a core dump 🤯

[Radonirinaunimi]

Thanks a lot @felixhekhorn for this!

1. somehow `pylint` doesn't seem to understand pineappl, see e.g. [here](https://github.com/NNPDF/pinefarm/actions/runs/17730698648/job/50381068173#step:10:30) . As a temporary fix I simply deactivate it, but this is surely not the right solution. I can reproduce that error locally, but I can also do `poetry run python -c "import pineappl; help(pineappl.grid.Grid.read)"` and get something sane back. What is the problem?

This is always a problem with compiled extensions and dynamically generated modules. PyO3 doesn't generate static files but compiled libraries. The problem is that pylint checks module hierarchies statically and therefore cannot understand the module structure, classes, and functions $-$ which in this case only exists at runtime.

The solutions are basically:

• instruct pylint to ignore the pineappl modules
• create .pyi stub files at the level of PineAPPL (Feature request: Generate stub files along during compile PyO3/pyo3#510); this is the most robust solution but as usual requires time, although crate like https://github.com/Jij-Inc/pyo3-stub-gen might help automating the generation

For the sake of not delaying this PR, perhaps we can live with the first option for the time being and implement the correct fix soon later.

[felixhekhorn]

The solutions are basically:

* instruct `pylint` to ignore the `pineappl` modules

you're right of course - stupid me; done in 4d7c8c3

pylint still correctly complaints about vrap

[Radonirinaunimi]

Apologies, I haven't had the chance to look into this yet (last week was very hectic). I will try to do so either later today or tomorrow.

[Radonirinaunimi]

This could be merged so that people could consistently produce grids/fks from this. I would address the Cyclic in another PR (as main is plagued with same warnings anyway).

[felixhekhorn]

@Radonirinaunimi actually, I just noticed this change: I guess line 195 is no longer needed and so is line 190? (i.e. temporarily writing to disk)