Overlap matrix for CC-PVDZ does not look correct

[xdvriend]

Describe the bug When retrieving the overlap matrix from GQCP in the cc-pvdz basis set, the resulting overlap matrix does not look as expected. It does not have ones on the diagonal.

Other basis sets (STO-3G, 6-31G,...) do not have this issue as far as we know.

To Reproduce H2 ccpvdz Overlap.txt

Expected behavior We expect the overlap matrix to look like this (bohr):

Ångstrom looks like this:

[guacke]

Current tests are only implemented for STO-3G and 6-31G. There is a discrepancy between the cc-pVDZ basis set included in the version of Libint included in the Docker container (see /usr/local/miniconda3/share/libint/2.6.0/basis/cc-pvdz.g94 (which is the same as the one used by Psi4))

H     0 
S   3   1.00
     13.0100000              0.0196850        
      1.9620000              0.1379770        
      0.4446000              0.4781480        
S   1   1.00
      0.1220000              1.0000000        
P   1   1.00
      0.7270000              1.0000000 

and the current EMSL basis set:

H     0 
S   4   1.00
     13.0100000              0.0196850        
      1.9620000              0.1379770        
      0.4446000              0.4781480 
      0.1220000              0.5012400
S   1   1.00
      0.1220000              1.0000000        
P   1   1.00
      0.7270000              1.0000000 

Using the latter basis set leads to ones along the diagonal, but to differing values off-diagonal.

[guacke]

Related to https://github.com/jjgoings/McMurchie-Davidson/issues/1 and https://github.com/theochem/horton/issues/39.