0.9.72

A few new features, and several bug fixes and improvements:

[Added or changed features]

• XAS Viewer (a naming demonstrating that Matt is poor at naming things) has been renamed "Larix".
• Structure information (from CIFs or data from the Materials Project) can now be converted to inputs for FDMNES. (PR #458). See the Jupyter Notebook example at https://github.com/xraypy/xraylarch/blob/master/examples/Jupyter/struct2xas_examples.ipynb. This should be viewed as ongoing work, but real progress has been made here, thanks to @beatrizfoschi @maurov @Ameyanagi @mretegan
• Binary Installers have been updated for all platforms, and are now based on Python 3.11.5.
• Significant improvements to Larch_XRD1D to read calibration files, integrate XRD images, display 1D XRD patterns, save XY outputs, and compare XRD patterns to crystal structures from the AMCSD database.
• Improvements in handling Bliss data, especially from ESRF BM16.
• Some improvements in Plotly plotting.
• Several cleanups and purging of the legacy _larch argument for many python functions.
• Some cleanups of XRF fitting results and generating XRF fitting scripts.
• Feffit reports now use floating-point formatting with strings of length 10 (was 11, led to many 1-digit diffs in outputs).

[Bug Fixes, Code improvements]

• fix for setting e0 with very badly formatted data.
• layout tweaks for Larix/XAS Viewer, especially for Linux/Gtk3.
• better checking of formula parsing for CIF formulas
• more gracefully closing AMCSD sqlite, avoiding crashes.
• more fixes to energy calibration.
• better defaults for normalization parameters when reading in data.
• several fixes in reading and working with Session Files and saved Feff calculations and Feffit results.
• better detection of non-default values for kstep in EXAFS routines.
• more uniform use of a single mkdir function.