Wketchum/genie generator hep mc3 refactor

[wesketchum]

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

What are the issues that this addresses?

This PR enables GENIE simulation that include the following:

• A HepMC3GenEvent object to add to the LDMX event bus, inheriting from the HepMC3::GenEvent.
• A EventWeights object to add to the LDMX event bus, for use in event reweighting studies.
• A GenieGenerator generator that interfaces with the GENIE neutrino generator to run lepton scattering models. Currently the generator is limited to starting interactions at the target. It uses a HepMC3GenEvent as an interface to the event bus.
• A GenieReweightProducer that can run GENIE's Reweight functionality, to reweight events based on turns of systematic knobs. It reads in HepMC3GenEvent recrods, recreates the GENIE internal event record based on that, and then runs reweighting.
• A GenieTruthDQM module that extracts truth GENIE interaction information from the HepMC3GenEvent metadata, and stores it in a flat tree.

Check List

(note, this requires use of the updated docker image described by this PR: LDMX-Software/dev-build-context#105)

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

An example ldmxsw config is here: https://github.com/wesketchum/LDMX_eN_GENIE/blob/main/ldmxsw_configs/genie_sim.py. From the output of that, one can look at the genie_truth tree from the DQM. Attached is a plot of the initial and final lepton energy for a 4 GeV run.

Additionally, to see the event reweighting working, here is a screenshot of some screen debug outputs that show weights applied per event.

[tomeichlersmith]

As a first pass, I just have some big comments

• Please review the coding rules, we solidified the format of variable names and such which you should apply here.
• Is ldmx::HepMC3GenEvent necessary? If you just need helper functions, I would prefer you to write those helper function somewhere else rather than derive a new event data type.
• Instead of using if (verbosity ...) checks, please use ldmx_log. I know we haven't transitioned other parts of SimCore to the logging system, but I would really like to avoid adding new code using the old method. It looks like your verbosity of 1 corresponds to the debug level.
• Does this compile with ldmx/dev:4.2.2? I just want to update the compatibility table if it doesn't.