User notes on the validity of two transformations

[SSL32081]

The two transformations, DistanceToSNRWeightedDistanceTransform and GeocentricArrivalPhaseToDetectorArrivalPhaseTransform, are designed to reduce parameter degeneracies, followed from arXiv:2207.03508.
These transforms assume the waveform consists of only the dominant quadrupolar mode, which is implicitly used in the two scaling factors: _calc_R_dets and _calc_R_det_arg. However, this assumption is invalid for most of the state-of-the-art waveform approximants, e.g. IMRPhenomXPHM and NRSur7dq4, with includes harmonics up to $\ell \leqslant 4$.

This fact was also noted by the authors in Sec. IV B, quote:

The symmetries in Section IV A are only approximate, since the detectors are not perfectly aligned, and general waveform models have additional effects like orbital precession and higher harmonics that are not modeled by Eq. (7).

Therefore, the documentation should include a note for users to use at their discretion or even raise a warning in the RunManager if they are used in conjunction with incompatible waveforms.
This is also a chance to update the documentation to include a reference to that article.

[ThibeauWouters]

This is certainly good to note in the documentation although not so urgent for now as eg IMRPhenomXPHM is not yet supported in Jim