Manifold Forward Euler should be implemented

[kellertuer]

...and for a while we should remove the ManifoldLie-Euler.

cc @mateuszbaran @sjacobsson

[mateuszbaran]

IIRC ManifoldEuler is the forward Euler on manifolds? And why should we remove ManifoldLie-Euler? This library can just depend on LieGroups.jl.

[kellertuer]

Sure, but the forward Euler on Manifolds currently does not exist here.

I was thinking of maybe doing a separate LieGroupsDiffEq, or do you think that is too much diversification? Then sure, we can also adapt the code here to use Lie groups in the corresponding places.

[kellertuer]

Ah, I missed we have ManifoldEuler, not sure how I missed it, but then this can be closed.

Only that we have to adapt to 0.11 and maybe also depend on LieGroups.jl then here (or do a separate LieGroupsDiffEq - for visibility . some time)

[mateuszbaran]

I think this package is too small for making such split. And currently I think the non-Lie group solvers are not particularly useful (I expect that in basically every problem either Lie groups solvers or solving in charts would be faster).