GitLab

TODO: Still need to understand why/what phase shift is needed.

Also removed some printed warnings in the internal waveform
generation functions.

- The CBC templates held in the models after the burnin were still zero, now
  they hold the actual burnin templates so that the MCMC can start with
  CBC templates subtracted from glitch/signal residuals

- Also moved the dtimes convention conversion to IN the projection routine
  so that we don't have to repeat it everywhere. The swapTimesFlag tells
  the routine which convention to use (=1 when called from BW, =0 when
  called internally from BayesCBC routines).