Fix the bug in the disentanglement step with SAWFs

[VVitale]

The bug was introduced in the following commit 5539b3c

This should now fix it and also works for parallel execution.

Fixes #314

[codecov]

Codecov Report

Merging #315 into develop will increase coverage by 0.60%.
The diff coverage is n/a.

@@             Coverage Diff             @@
##           develop     #315      +/-   ##
===========================================
+ Coverage    65.78%   66.39%   +0.60%     
===========================================
  Files           29       29              
  Lines        17957    17979      +22     
===========================================
+ Hits         11813    11937     +124     
+ Misses        6144     6042     -102     

Impacted Files	Coverage Δ
src/parameters.F90	83.05% <0.00%> (+0.11%)	⬆️
src/disentangle.F90	81.00% <0.00%> (+1.09%)	⬆️
src/sitesym.F90	93.97% <0.00%> (+30.59%)	⬆️

---

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 6386cfe...6f2432a. Read the comment docs.

[jryates]

There is an error in parallel which causes the test suite to fail. symmetrize_eps is not broadcast off the root node - hence if a none default value is set in the param file different nodes will apply the symmetry constraints to a different tolerance which propagates to give a different overall spread.

The fix is to add the following around line 6316 of parameters.f90
call comms_bcast(symmetrize_eps ,1)

In addition the testsuite will fail on OSX because of the line is jobconfig
program = WANNIER90_DISENTANGLEMENT_SAWFs
If we replace the final s with capital S it is ok