Monoclinic structure not stable under Niggli reduction
Our definition of monoclinic (from https://doi.org/10.1016/j.commatsci.2010.05.010) is not a good choice:
>>> from ase.build.tools import niggli_reduce_cell
>>> from ase.geometry import cell_to_cellpar
>>> cell1 = [[1,0,0],[0,1,0],[0,0.2,1]]
>>> cell2, _ = niggli_reduce_cell(cell1)
>>> cell2
array([[ 1.00000000e+00, 0.00000000e+00, 0.00000000e+00],
[ 6.12323400e-17, 1.00000000e+00, 0.00000000e+00],
[ -2.00000000e-01, 7.46914473e-17, 1.00000000e+00]])
>>> cell_to_cellpar(cell1)
array([ 1. , 1. , 1.0198039 , 78.69006753,
90. , 90. ])
>>> cell_to_cellpar(cell2)
array([ 1. , 1. , 1.0198039 , 90. ,
101.30993247, 90. ])