Fix lineage algo - RadioactiveDecayBase

[cfuselli]

With this PR we fix some issues with the treatment of RadioactiveDecayBase (RDB) interactions. I realised that the results of simulating one gamma and simulating the gamma from Xe131 gave very different results. That is not good. Now they both give similar results.

Specifically, we had to add special treatment for gammas emitted from excited nuclei that are treated by RDB library. Low energy gammas will not produce a gamma-phot/compt interaction in G4 just like hig energy gammas, but just some electron tracks and some gamma tracks will be produced. Now, we decide based on the secondaries what yield to assign to a certain cluster. The modifications match perfectly the distributions of Xe129 and Xe133 ( they were completely off before this PR ) and also match much better Pb212.

Details of changes:

• If there is "[" in the particle type ( like Xe129[236.140] ) - we pass also the secondaries to the classification. We need them to understand if to give a gamma or beta yield.
• Added get_particle_secondaries function to get first order daughters of a particle
• classification with secondaries: if there are only gammas -> it's a beta yield compton. If there are also electrons, it's probably a low energy gamma, so gets a gamma yield.
• lineage broken: brake if creaproc and edproc are both RDB. Intermediate step.
• modify ion lineage broken. before: brake if lineage is 6, now checks if ion.
• added: if lineage is a gamma yield, never brake anymore. We don't want to split photoabsorb.
• added: 0.1cm threshold specially for brem.

Xe129 - 39keV gamma

It makes some electron tracks and some gamma-phots. We want to keep them all together into a one gamma cluster.

Xe129 - 196keV gamma

It makes some electron tracks. Again, we want to keep them all together into a one gamma cluster.

Bi214 - 609keV gamma

It makes a real gamma. The gamma produces two compton clusters ( beta yield ) and one photoabsorbtion ( gamma yield )

Summary of clustering for different situations

Left: NEW clustering Right: OLD clustering

Gamma simulations 163keV in TPC

Expect just one gamma cluster, sometimes compton + gamma

Pb212

It's gammas+beta most of the times, with 238keV gamma

Xe131

It's one 163keV gamma

Xe129

It's two different gammas

BiPo

there is gamma+beta and then alpha

[cfuselli]

As you can see in the BiPo comparison, there was a difference that I did not notice between the A and Z values of the alpha cluster of Po214. It should be 4-2 and it was getting 214-84. This has an impact on the yields. I realised the only problem was that we were not splitting anymore for alpha types, adding that back all the rest is the same (checked for all the cases) but the alpha cluster now gets 4-2 for A-Z as it should.

[cfuselli]

Btw I am still discussing with Micha about this PR and we are still not sure how we want to treat those cases.

[cfuselli]

With the latest commit we do a bit of "dietrofront" on the treatment of RDB decays. I will update the description of the PR asap.

New treatment involves:

• merge together all electrons coming from excited stated RDB decays, they are conversion electrons, and we give them a beta cluster.
• keep doing the "if gamma break lineage". when we have both conversion electrons and gammas, we want to let the gammas travel and get a compton (beta yield) or photabs (gamma yield) cluster.
• we remove the need to classify based on secondaries and all the cases can be handled with just a few lines, specifying to not break if there is a "[" in the parent and to assing an 8 if there is an "[" in the type.