Revisit magnetic structure inputs

[janssenhenning]

The MaX6.1 release features support for defining collinear and non-collinear structures in the inpgen more easily, by defining a magnetic moment for each atom position. This enables roundtripping these kinds of structures from FleurinpData to InpgenCalculation

Questions/Needed Changes:

• How to store the magnetic moments associated with a structure? This is also an open question in aiida-core (see e.g. https://github.com/aiidateam/aiida-core/issues/4866)
  • Suggestion: subclass of StructureData with added magnetic moments/directions
  • What are nice ways of defining a magnetic structure from scratch? How should the interface look?
• masci_tools.io.fleur_inpgen needs to support the new format of defining magnetic moments