- assign IDs randomly so that HDF5 mass models cannot be used with the
  wrong template banks:  make the odds of template IDs matching if the
  wrong files are mixed small.

before they start assigning likelihood-ratios and FARs

feature_extractor.py: fix issue with missing channels in feature production due to using trigger_time rather than buffer time as a timestamp

All of the changes necessary to build and run itacac instead of itac. Note that itac will no longer work after this commit

- don't use obsolete .get_*() methods

- obsolete method of SnglInspiralTable class.

- use ligo.lw.table's new .columnnamesreal to get true column names

- fix process_id column name in sngl_inspiral table

far.py: change initial guess for extinction model to 5 which splits the difference between its previous value of 1 and its current value of 10.  We found some offline cases where 10 wasn't so great and 1 wasn't so great online. Hopefully 5 is better. FIXME ditch this scipy routine.

- bump minimum required python-ligo-lw to 1.5.3 (but don't enable check for it, yet)

- add M4 code to check for a functioning minimum required version of
  python-ligo-lw.  not yet used

- add ligolw_no_ilwdchar converions

- seems an accident of history left me with a header file installed in my
  accounts everywhere that was allowing me to compile this code when nobody
  else could

- oops, I was zeroing the row count before returning

- sngl_inspiral's process column is "process:process_id"

- this converts gstlal_snglinspiral_array_from_file() to a wrapper of
  ezligolw isntead of lal's XML reading functions