bug fix

[gitpeterwind]

In a rotation of orbitals, when the transformation matrix U is zero, the output was just a copy of the input, instead of zero. U is zero in the calculation of the Helmholtz argument when canonical orbitals are used (?).

[codecov[bot]]

Codecov Report

Base: 57.24% // Head: 57.23% // Decreases project coverage by -0.02% :warning:

Coverage data is based on head (181cc21) compared to base (e2f0e7e). Patch coverage: 0.00% of modified lines in pull request are covered.

Additional details and impacted files

```diff @@ Coverage Diff @@ ## master #205 +/- ## ========================================== - Coverage 57.24% 57.23% -0.02% ========================================== Files 178 178 Lines 14705 14709 +4 ========================================== Hits 8418 8418 - Misses 6287 6291 +4 ``` | [Impacted Files](https://codecov.io/gh/MRChemSoft/mrcpp/pull/205?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=MRChemSoft) | Coverage Δ | | |---|---|---| | [src/utils/ComplexFunction.cpp](https://codecov.io/gh/MRChemSoft/mrcpp/pull/205?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=MRChemSoft#diff-c3JjL3V0aWxzL0NvbXBsZXhGdW5jdGlvbi5jcHA=) | `0.00% <0.00%> (ø)` | | Help us with your feedback. Take ten seconds to tell us [how you rate us](https://about.codecov.io/nps?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=MRChemSoft). Have a feature suggestion? [Share it here.](https://app.codecov.io/gh/feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=MRChemSoft)

:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.

[stigrj]

Yes, with canonical orbitals the rotation matrix becomes zero in the iterations where we actually diagonalize, since the diagonal is subtracted from the Fock matrix in the rotation F_ij - Lambda_ij. In iterations where we don't explicitly diagonalize, the rotation matrix is slightly perturbed from zero.