planning/timeline/timeline.2013.xls: updated for progress

bugfix: inputs/input.Makefile: $(import): except in a full-database import, errexit so that the import will stop on an error and not let it scroll by

added inputs/CVS/^taxon_observation.**.sample/, used for the extract. note that the column list is slightly different than for VegBank.

inputs/CVS/taxonObservation_/map.csv: removed taxonObservation_-- prefix from terms that do not need to be table-specific (like for VegBank)

fix: inputs/CVS/taxonObservation_/map.csv: plantConcept_ columns: synced input and output column names to their names in plantConcept_

fix: inputs/CVS/taxonObservation_/map.csv: plantConcept_ columns: synced input and output column names to their names in plantConcept_

inputs/CVS/plantConcept_/map.csv: removed plantConcept_-- prefix from terms that do not need to be table-specific (like for VegBank)

lib/sh/db.sh: pg_table_exists(): use `SELECT NULL` instead of `SELECT *` to avoid a long column list cluttering up the log output

lib/runscripts/table.run: table_make_install(): simplified the setting of $noclobber since there no longer needs to be a different command for when the log exists

bugfix: lib/runscripts/table.run: need to errexit the make target, so that errors in the SQL install scripts are not suppressed. this requires pre-checking if the table exists (using new pg_table_exists), so that the install target's errexit does not then need to be suppressed for cases when the table already exists.