Factor out tracking, move to digi based seeds in the CI tests

[tvami]

I am updating ldmx-sw, here are the details.

What are the issues that this addresses?

#1551 does too many things at the same time, so I'd like to break it up to sub-PRs --> in this one I factor out tracking, and move to digi based seeds in the CI tests --> we expect the tests to show degraded (more realistic) performance. In a next PR I'll change to the optimized parameters.

Resolves #1595

Check List

• I successfully compiled ldmx-sw with my developments
• I ran my developments and the following shows that they are successful.

[tvami]

Here are the plots that fail the CI
signal_recon_validation_plots.pdf
it_pileup_recon_validation_plots.pdf
inclusive_recon_validation_plots.pdf
ecal_pn_recon_validation_plots.pdf

I'll concentrate on the inclusive for now:

• we seem to be having less track in general, which makes sense since now we use the digi seeds, not the truth seeds
• the quality of these tracks is worse than before, starting with the fact that we get more tracks per event: fig 4
• fig 8 - 14: in the trunk with truth seeds dont seem to have duplicates, now we do, which makes sense, that's why we have the ambiguity resolution (which is not in these plots)
• fig 27 - 32 is similar story
• the rest just shows degraded performance as well

@bloodyyugo @EBerzin I'd like to have this merged, so that we can update the gold files, and then I'll include the optimized parameters from @EBerzin in a next PR (which by that time can include the amb resolution)

[EBerzin]

The changes in tracking performance look consistent with what I saw when initially switching to reconstructed seeds from truth seeds.