GW and BSE method normalizer

Method normalization related with #1316 (closed).

Main changes:

• Added ExcitedState in results. Base class for results.method GW and BSE.
• Added method normalization for BSE and DFT+BSE, as well as generalizing GW and DFT+GW.

List of changes:

• Grouping GW and BSE as much as possible into base classes with names ExcitedState. This will be further useful when parsing TDDFT calculations.
• MethodNormalizer has been changed: new defined abstract class for ElectronicMethod, with the method to_simulation. This populates results.simulation.method for each specific electronic technique we currently have (DFT, GW/BSE, Projection, DMFT).
• Included BSE FilterMenu for the GUI under GW. This contains both BSE.type and BSE.solver as well as DFT+BSE and DFT+GW+BSE¹ metainfo (XC functionals, basis set names, GW Hedin type).

---

1. Thought there is currently not this DFT+GW+BSE workflow being defined or parsed, this is in my TODO list. For now I kept it as a possibility in the metainfo, but will revisit it soon. ↩