Reading geometry.in with fractional coordinates incorrect

[mondracek]

The input from an FHI-AIMS geometry.in file is not read correctly if the atomic coordinates are specified as fractional coordinates (using the atom_frac keyword); see e.g. the attached geometry.in.txt file here:
geometry.in.txt](https://github.com/user-attachments/files/16351689/geometry.in.txt)
The attached file contains a single layer of FeCl2, the elementary surface unit cell being repeated 5x5. So the structure should appear as a homogeneous hexagonal layer in AFM images (simulated using the Lennard-Jones potentials), but it does not.
I already know what the problem is and how to solve it (essentially a wrong order of matrix multiplication inside the loadGeometryIN function in ppafm/io.py), so I will assign myself and create the PR to resolve the issue.