Spin for valence calculations

[jtv3]

We need to add/check on spin support for all aspects of the code work flow.

0. Underlying vector operations in OCEAN_psi [x]
1. Energies [Das]
2. Wavefunctions (u2.dat) [Das]
3. Transition Matrix ElementsTransition [Das] a. For UV/optical b. For RIXS [x]
4. BSE a. H_0 (energy part) [x] b. Bubbles [x] c. Ladders [x]
5. PREP and DFT [jtv] a. umklapp b. psi's c. u2.dat d. enkfile e. range
6. OBFs [jtv]
7. Spin prefactor in opcons?

[jtv3]

Need to also add the ability to run a spin BSE on top of a nspin=1 DFT base. This would show up for instance in L23 RIXS of a non-magnetic system.

Proposing we add valence_ham_spin as an option in OCEAN_system:

1. Can be 1 or 2
2. Must be 2 if nspn = 2
3. If RIXS and not K-edge automatically 2
4. Undocumented, optional file for devs "force_val_ham_spin.ipt" ?

[jtv3]

From PRB 62, 4927 (2000). For loop structure of valence-level exciton valence spins are the outer loop, conduction spins are the inner loop.

The structure of the core-level exciton is: core-level spin, core-level m_l, conduction/valence spin. These two are more or less consistent with each other. The spin of the hole state is the outermost loop.

[jtv3]

Tests to run (@dasc01 feel free to add/subtract)

1. Si valence DFT spin=1, BSE spin=2
2. Si valence DFT spin=2, BSE spin=2
3. O2
4. Hematite
5. Check K-edge RIXS w/ BSE spin=2. The pnorm's will show if the spins are read in correctly.