Create Arbitrary Weight Function Option for GROUPR

[eitan-weinstein]

Similarly to the efforts of #285 which opened up ALARAJOYWrapper to take in any arbitrary group structure, the same should be done for GROUPR weighting functions. Not only would this be valuable in giving users more flexibility in their choice of how they want to prepare their data, but it would also allow us to prepare data even more similarly to FISPACT-II's PREPRO-processed groupwise nuclear data.

Related to the aims of #311, while I was more closely inspecting the contents and formatting of FISPACT-II's TENDL-2017 data, I noticed that the listed weight function was 1/E, rather than the VITAMIN-E piecewise function that is currently employed as the only option for ALARAJOY preprocessing. I found this result to be surprising, given that the stated groupwise processing for FENDL3 is explicitly weighted by VITAMIN-E (Schnabel et al., 2024). Therefore, for the purposes of producing a canonical ALARA library for FENDL3.2x should still be done using VITAMIN-E, but to best compare against FISPACT-II, a 1/E weighted library should also be produced.

[eitan-weinstein]

This discovery that FISPACT-II's processed TENDL data only uses a 1/E weighting function, whereas ALARA is capable of using VITAMIN-E (regardless of whether that is the only available option), adds an extra layer of value to this project. Ostensibly, someone could run PREPRO themselves over a TENDL data set (i.e. something like an "ALARAPRO" rather than "ALARAJOY" ...), but much like the process of this project, doing so would require a comparable effort, and is not something that is publicly available, to the best of my knowledge.