Rapidvertex

[gcowan]

Have a look and let me know what you think. Still need something for the SV resolution and someway to distinguish between PV and SV (derived class?). Logic for smearing different parameters (momentum, FD, IP) is not very unified, but we might want to sort that out later in a different branch.

[dcraik]

This branch is currently crashing a couple of the validation tests (https://gitlab.cern.ch/dcraik/RapidSim/pipelines).

• changes in ea9481ea13de061adba4e5339c55a9a3a7e91b53 cause Bs2DKpi to crash (can't see what's causing this one)

• changes in 27fbd37dec6c42028d386ab642c5ec3dc2c02834 cause Ds2KKpi to crash (most likely a null pointer to a RapidVertex object - we could fix this just by protecting each call with a check for a null pointer but I think everything should have an origin vertex so this is an indication that there's something unexpected happening elsewhere)

I've reverted some mass-indentation changes to make the changed files page easier to parse.

[dcraik]

Fixed some memory leaks that were upsetting the continuous integration and added vetex functionality for EvtGen-decayed particles.

Good to merge.