Dear pyscf developers,

I am using the selected CI module in pyscf (scisolver.kernel) to calculate a model system using fcidump.dat. However, I have encountered a strange problem that different version of pyscf gives different result, and the older version pyscf 2.0.1 gives a reasonable result but the newest pyscf seems to have a problem.

For example, I would like to solve for the excited state determinants in the HF basis, the result I obtain using pyscf 2.0.1 for the first excited state has the following determinants:
['00du22', '00d22u', '0d02u2', 'd00u22', 'd0022u', '00ud22', '00u22d', '0u02d2', 'u00d22', 'u0022d']
( the right 3 digits are the occupied HF orbitals, and the left 3 digits are the unoccupied HF orbitals)
There coefficients are:
[-0.7287080988581449, -0.1880852459026554, 0.3761704918053624, -0.18808524590265738, -0.4603626608047605, 0.15780358673095995, 0.040730336 963572644, -0.08146067392731682, 0.04073033696356988, 0.09969270162612504]

However, if I use the newest pyscf version is:
['00ud22', '00u22d', '0u02d2', 'u00d22', 'u0022d']
with the coefficients:
[-0.6175627497118902, -0.20410761294983304, -0.4082152260094989, -0.20410761294982663, -0.6071710995215007]

so we can see the newest version only considers exciting spin up electron to the unoccupied HF orbitals, but the old version considered both. I used exactly the same code for the two runs. Could I kindly ask if I need to specify extra variables in the newest version of pyscf to let it consider both spin up and spin down electrons in the excitation? Or is there any problem with the newest version of pyscf? Please let me know if I need to provide anything further such as the relevant part of the code. Thank you very much for your help!

My fcidump.dat is following:
&FCI NORB= 6,NELEC= 6, MS2=0,
ORBSYM=0,0,0,0,0,0
ISYM=1,
&END
0.300000 1 1 1 1
0.300000 2 2 2 2
0.300000 3 3 3 3
0.240000 4 4 4 4
0.240000 5 5 5 5
0.240000 6 6 6 6
0.600000 1 1 0 0
0.600000 2 2 0 0
0.600000 3 3 0 0
-1.040000 4 4 0 0
-1.040000 5 5 0 0
-1.040000 6 6 0 0
-0.100000 1 2 0 0
-0.100000 2 3 0 0
-0.100000 4 5 0 0
-0.100000 5 6 0 0
-0.100000 2 1 0 0
-0.100000 3 2 0 0
-0.100000 5 4 0 0
-0.100000 6 5 0 0
0.100000 4 4 1 1
0.100000 5 5 2 2
0.100000 6 6 3 3

Best,
Fangzhou Zhao