Kinematic Coverage (x-q) plot (will be) broken

[scarlehoff]

The x-q2 plot depended on the exact order of the kinematic variables to work and, together with that, the adequate kinematic transformation.

Since the values of x/q for a given dataset are not going to change, I suggest we should include them as part of the kinematics.yaml so that all kinematics.yaml have a coverage_Q2 and coverage_x entries (or just Q2 / x when it makes sense and it is not an approximation, like with DIS).

By doing this:

1. We have the explicit calculation in the filter.py file
2. We don't need kinematics_override, we save all kinematics we need.

Sadly (@t7phy @comane) this means you will need to update the datasets you already implemented so that they can be used with the new commondata.

@Radonirinaunimi the DY datasets seems to be mostly working, I guess you followed the same order they had before.

Note that this is something that affects both the ported commondata but also new datasets that you might want to implement. Without these variables it won't be possible to create the kinematic coverage plot.

A second option is to add a new key to the metadata files, similar to the kinematic_override but not making it explicit and only for the xQ2 mapping. So we have a key for mapping_xq and we add the relevant functions for that. This is with the idea that kinematics_override will eventually be only for legacy datasets.

[Radonirinaunimi]

Yes, I did follow exactly the previous ordering.