lib/sql_io.py: automatic handling of input/output column type mismatches: also do this for identifying columns, which first cause an error in a join in sql.distinct_table() rather than in the main insert (and thus were not handled by the existing error handling). previously, the user would have had to manually cast the input column in postprocess.sql. this involves getting handle_MissingCastException() to update join_cols as well as mapping.

lib/sql.py: distinct_table(): don't cache the table creation, because this prevents different distinct_tables from being created for the same input table

lib/sql.py: create_table(), copy_table_struct(): support custom query options, such as cacheable

lib/sql_io.py: put_table(): main loop MissingCastException handler: factored out into nested function so that it can also be used elsewhere

lib/sql.py: parse_exception(): parse "operator does not exist" errors as MissingCastExceptions (these appear when a staging table column is of the wrong type)

fix: lib/Parser.py: syntax_err(): error message: lightened the shade of gray so the boundary between the strings before and after the error would be clearer

lib/runscripts/in_datasrc_dir.run: datasrc_make(): use `if remaking ...` instead of accessing $_remake manually, for clarity

bugfix: lib/runscripts/in_datasrc_dir.run: datasrc_make(): use set_make_vars and $_remake as required in lib/sh/make.sh `remaking`

lib/sh/make.sh: remaking alias: documented that you MUST use set_make_vars at the beginning of any function that uses this, so that $_remake is properly set to $remake and not left at its previous value

moved everything into /trunk/ to create the standard svn layout, for use with tools that require this (eg. git-svn). IMPORTANT: do NOT do an `svn up`. instead, re-use your working copy's existing files with `svn switch` (http://svnbook.red-bean.com/en/1.6/svn.ref.svn.c.switch.html).