Add Di-Photon Performance Benchmark

[peter-mckeown]

BEGINRELEASENOTES

• Standalone python script createMCParticle_two_photons.py for creating an edm4hep input file to ddsim
• A Gaudi algorithm that processes the output of reconstruction and produces performance plots as a function of Di-photon separation
• Documentation in a README

ENDRELEASENOTES

This PR adds a Di-Photon benchmark, and has been ported to Gaudi/edm4hep from an original Marlin/LCIO workflow written together with @akorol98. It was written for testing fast simulation models, but was based on benchmarks originally used in the development of Pandora see (Chapter 5, B. Xu PhD Thesis), so we hope will also be useful for reconstruction algorithm development.

• The file createMCParticle_two_photons.py writes an edm4hep file that acts as an input file for ddsim. The readme includes an example for generating two photons with variable separation at the face of the ILD ECAL with this script. This allows the user to target specific regions of the detector as desired. The photons are created as if they flew from the IP, as the particle ID Pandora assigns to the PFO is sensitive to this.
• A Gaudi algorithm produces ROOT plots for the average number of PFOs reconstructed for varying photon separations. The true photon separation is obtained from the MC particles linked to the PFO(s). The fractional fragment energy (i.e. fraction of the reconstructed energy not contained in the two leading PFOs) is also recorded.