feat: use algorithms properties in CalorimeterClusterRecoCoG

[wdconinc]

Briefly, what does this PR introduce?

This PR converts CalorimeterClusterRecoCoG to using properties defined as algorithms::Property.

TODO:

• Unify factories #865 (comment)

Cherry-picked d498abe from #1161.

What kind of change does this PR introduce?

• Bug fix (issue #__)
• New feature (issue #__)
• Documentation update
• Other: __

Please check if this PR fulfills the following:

• Tests for the changes have been added
• Documentation has been added / updated
• Changes have been communicated to collaborators

Does this PR introduce breaking changes? What changes might users need to make to their code?

No.

Does this PR change default behavior?

No.

[github-actions]

Capybara summary for PR 1154

• rec_dis_10x100_minQ2=1000_craterlake
• rec_dis_10x100_minQ2=1_craterlake
• rec_dis_18x275_minQ2=1000_craterlake
• rec_dis_18x275_minQ2=1_craterlake
• rec_dis_5x41_minQ2=1_craterlake
• rec_e_1GeV_20GeV_craterlake
• rec_pi_1GeV_20GeV_craterlake

[wdconinc]

Cool, no diffs at https://github.com/eic/EICrecon/actions/runs/7015508837/job/19085214064?pr=1154#step:11:104

[wdconinc]

This is ready for review and merging.

[veprbl]

This worked out nice. Having a map of properties helped a bunch.

[veprbl]

The full error message is 4 gigabytes, but can be summarized as:

In file included from /home/runner/work/EICrecon/EICrecon/src/detectors/EEMC/EEMC.cc:13:
/home/runner/work/EICrecon/EICrecon/src/factories/calorimetry/CalorimeterClusterRecoCoG_factoryT.h:66:67: error: 'key' in capture list does not name a variable
            [app, &m_algo = m_algo, &m_log = m_log, param_prefix, key](auto&& val) {
                                                                  ^

[wdconinc]

There are new LSAN errors in CI.

Actually, this is more curious (and tl;dr no memory leaks). What happened is that within a pipeline one job checked the geometry version in eic-shell and then decided it didn't need to regenerate the simulation file (cache hit). Then the reconstruction job started in the next job, in its own eic-shell, and it had a newer geometry version due to the cvmfs sync, so it continued to run with the older simulation geometry input file but against newer geometry in memory.

The geometry difference was https://github.com/eic/epic/pull/603/files -> i.e. r,phi segmentation to x,y segmentation. That probably caused many warnings about missing cellIDs (including MaxCellIDerrors failures), and memory errors are probably associated with that. Probably ok to let this slide...

Last checks indicate no memory issues anymore: https://github.com/eic/EICrecon/actions/runs/7038493136/job/19156610394?pr=1154#step:8:2091

[wdconinc]

And no diffs, e.g. https://github.com/eic/EICrecon/actions/runs/7050567792/job/19192283374?pr=1154#step:11:106. After 9ee8a85 we should get a clean clang-tidy-iwyu check too. And the eicmkplugin fix is in the merge queue.

So, good for review and merge, as far as I'm concerned.

[wdconinc]

This PR has become a bit more difficult now that #865
has been merged. Instead of looping over the algorithm's properties and parsing using Jana itself, we probably need to loop over the algorithm's properties inside JOmniFactory's constructor if constexpr the algorithm has such properties, and then store ParameterRefs in a container of ParameterBase smart pointers. The ParameterRefs will register themselves and get parsed and we don't need to access the container anymore.

[wdconinc]

Deferred to #1165