Adding fix for ITk SP production
This MR attempts to fix the lack of sp in the ITk strip endcap. The reason why the current code doesn't work is due to a different assumption of the orientation of the local reference frame of the SCT endcap wrt the ITk strip endcap. For the current SCT endcap, back-to-back modules have same-direction normal vectors, hence the local frames on the two modules are oriented in the same way. To check for compatible clusters on the two modules, a simple difference of the local-x coordinates can be evaluated. Instead, for the ITk strip detector, modules on the two sides have opposite normal vector, which implies a rotation around the strip direction of the local frame. This means, that close-by clusters can present opposite-sign local-x coordinate. As a consequence, the check described above doesn't work any more.
In the last commit, also a fix for the phi overlap space point formation is provided. Again the assumptions in the geometry description don't agree between ITk and ID.
The fix proposed, should preserve the Run2/3 behaviour.