[Wien] local rotation matrix in QTL program

[Masao ARAI]

Dear WIEN2k developers and others,

Thank you for a great package WIEN2k.

Recently, I encountered a strange problem in QTL program.
It may be related to the problem reported to this ML 
by Prof. M. W. Haverkort with the title  "new x and z axis in qtl".

I calculated pdos of distorted perovskite. 
But, the PDOS of d orbitals changed when I rotated local coordinate 180 degree.

Namely, I tried following two local coordinate.