Parameterized Kalman Filter gives state at first measurement

The parameterized Kalman Filter (!1693 (merged)) provides a state at the first measurement instead of the state closest to the beamline. This change in behavior affects many SMOG lines, which cut on the z position of the track state with the expectation that this is determined at the track's point of closest approach to the beamline. I think there are two things we should do:

1. Propagate the KF state to the beamline and save the state at the point of closest approach. This way the parameterized KF does not change the expected behavior of track states for line writers. (!1918 (merged))
2. Add Allen CI tests on SMOG with the parameterized KF. I can do this in the same MR as 1.
3. Remove the dependency on the position of the track state from lines, as this is not uniquely defined. E.g. we should use an explicit calculation of POCAz in SMOG lines. I think this should be a longer term goal, as it requires touching a lot more code.

FYI @dovombru @mveghel @leuecker @oboenteg @samarian