The primary need for sanity checks was on radius, which has now been built into BayesX. There is room for additional checking e.g. checking all input data has the correct dimensions.
Consistency checking is being set aside for now as the existing standalone script is entirely serviceable.
Masking support is partially implemented but won't automatically transform .reg files for now
Bring updated Python code and documentation into main