Thomson problem in OPTIM

This is what I have done to make the Thomson problem works in OPTIM:

1. copy Gthomson.f90 from GMIN and work out the Hessian

2. Fudge the number of atoms in fetchz.f

3. In geopt.f, NEXMODES = 1

4. In hybridmin.f, add SHIFTHGTH

5. Does the keyword reading in keywords.f

6. Edit potential.f, key.f90

7. Edit efol.f90, ZT(1) = .FALSE.

8. Add GTHOMSONT in myorient.f90 to advantage of the symmetry

7. Lots of edits in CONNECT/ncutils.f90.

8. In mindist.f, minpermdist.f90, newmindist.f90, recall if they are in the reduced representations. To do this I need to copy these subroutines and duplicate them with HK...f90. See Gthomson.f90

To do?

1. In OPTIM.F, SHIFTL(1-6) = SHIFTV, need to specify all 6?

2. It seems eigenvalue shifting is not really working - check bfgsts.f. I have to use NOHESS

3. Potential problem in ADDNEWMIN. Ignore error when D.LT.GEOMDIFFTOL