Double-pass image extractor

[HealthyPear]

Description:

This feature is relevant for issues #24, and in particular for issues #31 and #37 .
It has been already pushed to ctapipe (see cta-observatory/ctapipe#1215 for details).
This version complies with the conda-packaged version of ctapipe 0.7, so expect some (not performance-related) differences.

Caveats:

• during the tests for the comparison against CTA-MARS it has been necessary to hamper heavily with the code of ctapipe; for this reason a lot of classes and their methods needed to be overwritten
• also during said tests it was necessary to use all single-telescope images, feature which required to modify heavily the code of protopipe. These changes will be overwritten by the new DL1 writer, and they could also disrupt too much the next steps of the pipeline - for this reason I preferred to keep it simple and leave that part as original as possible. Protopipe will use this image extractor, but the (optionally) exported images will be only those which survive AND contribute to direction reconstruction
• together with the usual DL1 charge and pulse-times (for both passes!), this image extractor exports also a variable called calibration_status (only for the optional images.h5 file); such variable records if an image doesn't survive the 1st pass OR survives it but the image-width from the fit between the 2 passes is NaN; for the previous point, this is of course useless now (since only good images are exported), but I decided to keep it here if we decide to implement this functionality later on.

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (master@d76c0e1). Click here to learn what that means.
The diff coverage is 0%.

@@           Coverage Diff           @@
##             master    #48   +/-   ##
=======================================
  Coverage          ?   0.4%           
=======================================
  Files             ?     20           
  Lines             ?   2221           
  Branches          ?      0           
=======================================
  Hits              ?      9           
  Misses            ?   2212           
  Partials          ?      0

Impacted Files	Coverage Δ
protopipe/scripts/write_dl1.py	0% <0%> (ø)
protopipe/pipeline/event_preparer.py	0% <0%> (ø)

---

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 d76c0e1...4a07fcc. Read the comment docs.