Update ManifoldDiff.jl compat to 0.4 and CI

[mateuszbaran]

No description provided.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 93.19%. Comparing base (5d60ce5) to head (9f219f1).
Report is 7 commits behind head on main.

❗ There is a different number of reports uploaded between BASE (5d60ce5) and HEAD (9f219f1). Click for more details.

HEAD has 1 upload less than BASE

Flag	BASE (5d60ce5)	HEAD (9f219f1)
	4	3

Additional details and impacted files

@@             Coverage Diff             @@
##              main      #23      +/-   ##
===========================================
- Coverage   100.00%   93.19%   -6.81%     
===========================================
  Files           10       11       +1     
  Lines         1106     1191      +85     
===========================================
+ Hits          1106     1110       +4     
- Misses           0       81      +81     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[mateuszbaran]

@kellertuer I have no idea why building the tutorial gets stuck on CI. Could you try it on your computer? I still don't have a working quarto installation.

[kellertuer]

Hm, clear cache here, and maybe also let the examples run on 1.11?

[mateuszbaran]

There are no caches in this repository and the example should work fine on Julia 1.10.

[kellertuer]

Ok, I can try, but probably only tomorrow, just came home and first need something to eat.

[kellertuer]

Bumped all versions (Julia for documenter and quarto, quarto, version on the example manifest), my local version is on 15/21 in the nb that got stuck, a few seem to run a bit slow, yeah.

[kellertuer]

Yeah maybe the subsolver calls take too long on cell 20/21, but I have no clue how to directly debug that for now.
This week seems to become the week everything fails if this continues.

[mateuszbaran]

Maybe just setting a lower limit of iterations would be fine? At least as a temporary solution until you have time to investigate.

[kellertuer]

Will check, the current parameters are the ones from the paper, so it would be a pity to change that, but well, maybe that is what it is and it turns out nearly nothing is ever reproducible.