Minor Changes to DMRG(Block2)

[lcyyork]

Description

Some minor changes regarding to the new block2 interface.

• [x] Fix the read_active_wfn bug when multiple states are available.
• [x] Remove the MPS compressed intermediates for DSRG after the computation.
• [x] Fix the DMRG-SCF maxiter in the "nail-down" stage.
• [x] Read MPS compressed intermediates for LDSRG(2) computations.
• [x] Add Cholesky active orbitals for a trade-off of localizing orbitals while maintaining point-group symmetry. (Fast noniterative orbital localization for large molecules)

User Notes

• [x] Add Cholesky orbitals for localizing active orbitals.
• [x] Several options are added:
  • BLOCK2_INTERGRAL_CUTOFF: considered as zero if integrals are smaller than this threshold.
  • BLOCK2_INITIAL_GUESS_OCC: active occupation numbers (in Pitzer order) for DMRG initial guess. By default, we use random initial MPS. If not, one can specify this option in combination with BLOCK2_INITIAL_GUESS. My preliminary tests show the occupation bias (i.e., BLOCK2_INITIAL_GUESS) should be around 0.5.
  • DSRG_3RDM_BLOCK2_CPS_BOND_DIMENSION: bond dimension used to compute Ψ(N-1) = h_{pσ} (t) |Ψ (N)> = \sum_{uvw} t_{pw}^{uv} \sum_{σ1} w^+_{σ1} v_{σ1} u_{σ} |Ψ(N)>. My preliminary tests show this value should be at least three times the bond dimension of |Ψ(N)> to reach mEh accuracy for the 3-cumulant contribution in DSRG-MRPT2.

Checklist

• [x] Added/updated tests of new features and included a reference output.ref file
• [x] Removed comments in code and input files
• [x] Documented source code
• [x] Checked for redundant headers
• [x] Checked for consistency in the formatting of the output file
• [x] Documented new features in the manual
• [x] Ready to go!

[fevangelista]

I'd like @brianz98 to check this PR since he has also done work on the DMRG testing. It also looks like there is some cleanup that needs to be done. BTW, can you say more about the changes made in this PR? The nail-down stage code seems mostly commented out, IIRC.

[lcyyork]

I'd like @brianz98 to check this PR since he has also done work on the DMRG testing. It also looks like there is some cleanup that needs to be done. BTW, can you say more about the changes made in this PR? The nail-down stage code seems mostly commented out, IIRC.

I am sorry, but this PR is not ready for review.

[fevangelista]

Thanks everyone! Is this ready for merging now?

[lcyyork]

Thanks everyone! Is this ready for merging now?

I think yes.