fix: Use material before the first measurement in smoothing

[Corentin-Allaire]

This PR allow material situated before the first measurement (eg : beampipe) to be properly accounted for in the KF and CKF. This will fix issue #906

[codecov]

Codecov Report

Merging #956 (c3485ea) into main (0123060) will decrease coverage by 0.01%.
The diff coverage is 39.28%.

@@            Coverage Diff             @@
##             main     #956      +/-   ##
==========================================
- Coverage   48.47%   48.46%   -0.02%     
==========================================
  Files         334      334              
  Lines       17059    17067       +8     
  Branches     8069     8074       +5     
==========================================
+ Hits         8270     8271       +1     
- Misses       3088     3094       +6     
- Partials     5701     5702       +1     

Impacted Files	Coverage Δ
...de/Acts/TrackFinding/CombinatorialKalmanFilter.hpp	29.94% <29.41%> (-0.23%)	⬇️
Core/include/Acts/TrackFitting/KalmanFitter.hpp	37.83% <54.54%> (-0.23%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 0123060...c3485ea. Read the comment docs.

[paulgessinger]

Seems fine to me, but let's wait for @XiaocongAi's review.

[XiaocongAi]

Hi @Corentin-Allaire , thank you for the PR. I have only one comment.
In the CKF, shall we do the following:?

1. If we are going to run smoothing to get the fitted track parameters, we do the same as in KF, i.e. we create only material state before the first measuement and create measurement/hole/material after the first measurement;
2. If we are don't care about the track fitting, i.e. the CKF is used just for track finding, we only create measurement/hole state after the first measurement.

[Corentin-Allaire]

Hi @Corentin-Allaire , thank you for the PR. I have only one comment.
In the CKF, shall we do the following:?

1. If we are going to run smoothing to get the fitted track parameters, we do the same as in KF, i.e. we create only material state before the first measuement and create measurement/hole/material after the first measurement;

2. If we are don't care about the track fitting, i.e. the CKF is used just for track finding, we only create measurement/hole state after the first measurement.

I think this is a great idea, I merged your suggestion.

[paulgessinger]

Ok this seems fine now. Let's merge it.