Change IFU telluric correction to three recipes, model, calctrans, and correct

[hugobuddel]

From Wolfgang

Hugo, Gilles and I have discussed telluric correction approach for the
IFU mode. Currently it is implemented in two recipes, the
metis_ifu_telluric, dedicated to derive the transmission and flux
response function, and metis_ifu_calib, which applies both, the telluric
correction and the flux calibration.

The usual ESO approach is to have three telluric corr recipes for
molecfit (*_model, *_calctrans, *_tellcorr, aiming for the fitting step,
calculation of the entire transmission, and the application of the
latter, respectively), and an additional one for the flux calibration.

We think, a similar approach would be sensible for the IFU mode, also to
sync with the LSS. However, this would require some changes in the IFU +
EDPS workflows, which we should discuss in more detail (I guess with
Martin as main IFU wkf developer, right? Someone else?).

These changes require a certain amount of effort for implementing them
in the workflow and updating the DRLD. We therefore decided to keep the
structure as is for the mid-Feb release, and inform Lars & Lodo that
there will be changes.

I hope this approach is ok for you?