Inverted rotation axis for some data sets read by `FormatHDF5EigerNearlyNexus`

[dagewa]

These data sets from 3 different facilities:

• https://proteindiffraction.org/project/7brr/ - SSRF BL17U1
• https://proteindiffraction.org/project/step_covalent_9eez/ - CHESS 7B2
• https://proteindiffraction.org/project/8YS9_xray-data_8ys9/ - PAL/PLS 5C (4A)

all import with an inverted rotation axis. I do not know if this is common to all data sets from those beamlines.

[CV-GPhL]

These datasets are a typical example of changing HDF5 files without understanding coordinate systems, standards etc ... After all, a "proper" HDF5 dataset following the Gold Standard should have all axes defined consistently (within the McStas coordinate system) using the metadata items available internally.

• 7BRR and 8YS9 contain completely crippled HDF5 metadata: no /entry/sample/transformations and no /entry/instrument/detector/module/{fast,slow}_pixel_direction vectors.
• 9EEZ is weird in that it contains changing angle values for all axes: Phi (0.2 deg), Omega (0.2 deg), Chi (0.1 deg) and Kappa (0.1 deg) ... unlikely that was an experiment with 4 rotation axis changing at the same time ;-)

The fact that two of these datasets are still collected with 1 degree image width (CCD or even Image Plate thinking) on Eiger detectors in 2022 beggars belief.