small fix for SLWF+C

[hjunlee]

Dear all:

This commit does the following two things:

(1) enable the SLWF+C for all Wannier functions (WFs):
Currently, if the keyword of slwf_num is missing or slwf_num=num_wann is specified, selective_loc is set to .false. In this case, we have no way of performing the Wannierisation with centres fixed since the block of constraint is nested within the block of selective localisation.

I think that there might be some cases in which it is necessary to perform the SLWF+C for all WFs.

(2) very minor bug fix for printing the penalty function

Sincerely,

Hyungjun Lee

[codecov]

Codecov Report

Merging #287 into develop will decrease coverage by 0.07%.
The diff coverage is 100%.

@@             Coverage Diff             @@
##           develop     #287      +/-   ##
===========================================
- Coverage    65.65%   65.57%   -0.08%     
===========================================
  Files           29       29              
  Lines        17984    17910      -74     
===========================================
- Hits         11807    11745      -62     
+ Misses        6177     6165      -12

Impacted Files	Coverage Δ
src/parameters.F90	82.84% <100%> (+0.08%)	⬆️
src/wannierise.F90	81.72% <100%> (+0.05%)	⬆️
src/io.F90	58.08% <0%> (-12.27%)	⬇️
test-suite/library-mode-test/test_library.F90	64.7% <0%> (-0.35%)	⬇️
src/postw90/geninterp.F90	74.83% <0%> (-0.17%)	⬇️
src/postw90/postw90.F90	86.74% <0%> (-0.16%)	⬇️
src/comms.F90	52.63% <0%> (-0.16%)	⬇️
src/wannier_prog.F90	81.81% <0%> (-0.15%)	⬇️
src/plot.F90	49.45% <0%> (-0.06%)	⬇️
src/postw90/get_oper.F90	64.58% <0%> (-0.05%)	⬇️
... and 11 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c658e7e...eba560b. Read the comment docs.

[VVitale]

I believe there is a reason for this, which is the following: The sum of the centres of the Wannier function should be invariant modulo a lattice vector, therefore you cannot fix the centres of all Wannier functions to some target positions.
In the original paper by Marianetti et al. they never fix all centres but they only fix the centres of the selected WFs.

[hjunlee]

You are right. I will close this commit.