bugfix: /README.TXT: Full database import: To restart an aborted import for a specific table: added by_col=1

/README.TXT: Full database import: added steps to restart an aborted import for a specific table

inputs/input.Makefile: $(import): support restarting the import where it left off by setting continue=1. this is done by grepping the restart row out of the log file's last partition.

inputs/input.Makefile: added %/import_scrub, similar to import_scrub but just imports one table

bin/import_all: with_all import_scrub: documented that this step uses $by_col, so that users know to include by_col=1 when running this step separately

bin/import_all: use column-based import (by_col=1) by default, instead of requiring the user to explicitly specify it. instead turn it off explicitly (by_col=) for row-based import.

bugfix: /README.TXT: Full database import: To back up DB: after renaming current import to public: say to replace $version with the appropriate revision, because the $version env var should not be set (otherwise the backup will try to use a nonexistent import with the given revision #)

/README.TXT: Full database import: To back up DB: updated instructions to inline setting of $dump_opts, like in bin/import_all

bin/import_all: don't set $dump_opts until running the backup command that uses it, so that the user can run this backup command separately just by copying the line out of the script (without worrying about env vars that need to be set, other than $version which is visible outside the script)

inputs/U/Specimen/: translated some multi-column filters to postprocessing derived columns, using the steps at http://wiki.vegpath.org/Adding_new-style_import_to_a_datasource#Translating-filters-to-postprocessing-derived-columns