bin/import_all: `make -s`: removed no longer needed output_data=1

bugfix: bin/import_all: don't disable errexit because this prevents the program from being Ctrl-C'd. this functionality is no longer needed now that the README.TXT instructs to run bin/import_all in a subshell.

bin/import_all: removed functionality now provided by util.run

bin/import_all: converted to a runscript so it can use runscript functionality

bugfix: bin/make: don't run verbosity_compat until right before executing the external command, so that it doesn't mess up the logging mechanism. this is run automatically by command(), so there is no need to do anything here. note that logging bugs like these can now be troubleshooted much more easily with pst() to narrow down which functions could be causing the problem.

schemas/public_.sql: viewFullOccurrence_*: renamed to view_full_occurrence_* at Brian M's and Martha's request (e-mails from Martha on 2014-8-12 at 17:37PT, and from Brian M on 2014-8-13 at 16:21PT). note that this change has already been made on vegbiendev.

lib/tnrs.py single_tnrs_request(), bin/tnrs_client: use_tnrs_export: default to False because this mode uses incorrect selected matches (vegpath.org/issues/943), and the JSON mode that fixes this is now available

bin/tnrs_db: tnrs.tnrs_request() call: explicitly set use_tnrs_export=True so that this continues to work if the default value is changed

bin/tnrs_client: added env var to configure use_tnrs_export

bin/make_analytical_db: materialize viewFullOccurrence_individual_view instead of analytical_stem_view because analytical_stem_view is now generatable via a simple join onto viewFullOccurrence_individual_view. this avoids running into potential disk space constraints when materializing and backing up both tables (~50 GB/table * 2 tables * 2 copies (incl. the backup) = 200 GB, which is very close to the available disk space).