[catalogue] Added the tape states to the catalogue::getTapes() methods + TapeSearchCriteria + unit tests + refactoring

[catalogue] Added the tape lifecycle columns in the database + the method to modify the state of the tape

[cta-admin] Implemented cta-admin recycletf ls [--vid] [--fxid] to list content of the FILE_RECYCLE_LOG table

[catalogue] Storage class deletion modified to check if there are files on the FILE_RECYCLE_LOG table before deleting the storage class

[catalogue] TAPE_FILE.SUPERSEDED_BY_VID and TAPE_FILE.SUPERSEDED_BY_FSEQ are not used in any query anymore

Oracle DB now insert repacked file to the recycle log and deletes the associated TAPE_FILE entries. Postgres WIP

Adds fields for tape verification to Workflow message

This reverts commit 9ba6951d.

The orginal commit is being reverted because it intentionally
forced the "list contents of a tape" query to use indicies by
replacing table joins with sub-queries.  These sub-queuries
unfortunately turned out to be slower than the original query
when it was executed in a "clean" environment and the database
live memory cache was cold.  By "clean" environment I mean one
without PARALLEL enabled on the following two tables:

ARCHIVE_FILE
TAPE_FILE

PARALLEL is enabled on these two tables when a CASTOR to CTA
migration takes place.  It was left in place with the hope that
it would not interferre with the day to day running of CTA.  This
has now been proven to not be the case.

Corrected valgrind error for the DataTransferConfig::tapeLoadTimeout attribute not set in the unit tests

[cta-taped] Added configuration TapeLoadTimeout to set the timeout after which the tape mount is supposed to be failed