Reworked 3d conformal and sequencing

[JenHardt]

3d conformal now no longer errors out when used together with sequencing, the issue here was the inconsonant number of bixels in the stf, which is why the collapseStf function was added. Also changes to the sequencer were made to accomondate for example a beam with 0 weight vector.

this also reminded me that we should finish the work on the class based sequencers

[codecov]

Codecov Report

❌ Patch coverage is 73.04348% with 31 lines in your changes missing coverage. Please review.
✅ Project coverage is 54.05%. Comparing base (d2daafd) to head (a89221f).
⚠️ Report is 1 commits behind head on dev.

Files with missing lines	Patch %	Lines
matRad/optimization/matRad_collapseStf.m	0.00%	9 Missing ⚠️
matRad/sequencing/matRad_siochiLeafSequencing.m	69.23%	8 Missing ⚠️
matRad/sequencing/matRad_engelLeafSequencing.m	66.66%	5 Missing ⚠️
matRad/sequencing/matRad_xiaLeafSequencing.m	66.66%	5 Missing ⚠️
matRad/gui/widgets/matRad_WorkflowWidget.m	0.00%	2 Missing ⚠️
matRad/optimization/matRad_collapseDij.m	95.55%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##              dev     #885      +/-   ##
==========================================
+ Coverage   53.85%   54.05%   +0.20%     
==========================================
  Files         308      309       +1     
  Lines       19939    20007      +68     
==========================================
+ Hits        10738    10815      +77     
+ Misses       9201     9192       -9     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[github-actions]

Test Results

    3 files  ± 0      3 suites  ±0   45m 37s ⏱️ +16s
  325 tests + 6    325 ✅ + 6  0 💤 ±0  0 ❌ ±0 
1 071 runs  +18  1 066 ✅ +18  5 💤 ±0  0 ❌ ±0 

Results for commit a89221f. ± Comparison against base commit d2daafd.

♻️ This comment has been updated with latest results.

[wahln]

The tests are failing for some reason due to a missing "sum field".

[wahln]

There was a mistake in how the dij.numOf fields were interpreted which I fixed. I also used the opportunity to expand to collapse dijs to other modalities, which should work now. This affected some other tests and calculations which I fixed as well.

[wahln]

Some other remarks:

• For protons and other ions, the current collapsing means basically single field uniform mode. This might be also interesting to implement as a change in the GUI.
• For protons and other ions, it might also be interesting to collapse per ray (which somewhat simulates an irradiation with a ridge filter producing a SOBP from scattering). Maybe the collapseDij field could have a corresponding mode field (mode="beam" -> default, mode="ray").

Do you want to address this in this PR or in a following one?

[JenHardt]

Some other remarks:

• For protons and other ions, the current collapsing means basically single field uniform mode. This might be also interesting to implement as a change in the GUI.
• For protons and other ions, it might also be interesting to collapse per ray (which somewhat simulates an irradiation with a ridge filter producing a SOBP from scattering). Maybe the collapseDij field could have a corresponding mode field (mode="beam" -> default, mode="ray").

Do you want to address this in this PR or in a following one?

I think this can be adressed in this pull request, i am just a bit unsure how to best add this to the GUI, like how would we for example call the raywise mode for ions

[wahln]

The GUI thing could be secondary. We could for now just rename to 3D Conformal / SFUD. The ray mode does not need to be in the GUI for now.

BTW: Did a few updates for the tests to run through, hopefully