fix: enable photon vertex cheating by default, for MC info access

[c-dilks]

Briefly, what does this PR introduce?

enables cheatPhotonVertex for testing, so that drich-dev benchmarks will show the residual plots; note that this cheat mode does not impact PID, since it only allows knowledge of some MC to be read out, but not actually used. In other words, turning on this cheat mode does not actually "cheat" the PID... the results are the same when it is off.

This is in some sense a "fix", since we cannot otherwise check the residual plots.

Renames, including the enabled cheatPhotonVertex:

recon_irt.yaml -> recon_main.yaml, and is now used as the default
recon_irt_noise.yaml -> recon_noise.yaml

What kind of change does this PR introduce?

• [x] 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
• [x] Changes have been communicated to collaborators: @chchatte92

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

no

Does this PR change default behavior?

yes, makes residual benchmarks executed here in drich-dev useful

[chchatte92]

We need the mc info in two places.

• We need them in the residual computation.

• I am also thinking to plot also mc wavelength versus Reconstructed angles. The following plot is because when we will have noise injected one can check the differences and added noise will be independent of wavelength it we find this, this will give us confidence on added noise.

  For the first part though, I have though a way around. We compute average the refractive index once and pass it as a constant to estimate the theta_expected. Because the knowledge of the (n-1) will come either from a measurement or computing it from saturated photons' angle. After a while the mc wavelength plots can be omitted.

[c-dilks]

For now your changes can just go in reconstruction_benchmarks. This PR makes sure that when we run reconstruction_benchmarks from drich-dev, we can see everything we are used to seeing. This should hold us over until reconstruction_benchmarks is triggered by upstream PRs, or we migrate our benchmarks elsewhere (see Mattermost DMs).