Bug in LAW 3 Level Inelastic Scatter

[HunterBelanger]

Currently, this documentation states that for level inelastic scattering, LDAT(1) = (1+A)*abs(Q)/A, implying that LDAT(1) should always be positive. This is typically true as Q values are generally negative, but this is not the case when looking at excited states, which may have positive Q values. One such example is Am244m1. Looking at line 10361 in acefc.f90, NJOY actually writes LDAT(1) = (1+A)*(-Q)/A. This is an important distinction for error checking in ACE files, as what NJOY actually writes can indeed be negative for excited states.

I have a fix prepared for this, but I am waiting to submit the PR, as there are several things that are waiting to be merged, and the PDF will is likely causing enough merge conflicts already.