Added activable logs to the RetrieveQueueStatistics cache for debugging

An example value of sys.retrieve.error is as follows:

[itctabuild02] ~ > sudo eos attr ls /eos/dev/userfiles/testdir_1/test_1 | grep sys.retrieve.error
sys.retrieve.error="Garbage collected at 2019/09/09 16:27:26.609570"
[itctabuild02] ~ >

This can now be done without tearing down and re-creating the containers.

Added error messages in case of problems for easier diagnosis.

Conflicts:
	continuousintegration/docker/ctafrontend/cc7/opt/run/bin/ctaeos-mgm.sh

Conflicts:
	continuousintegration/orchestration/tests/repack_systemtest_wrapper.sh

Added report generation to repack_generate_report.sh so that it honours the repack "add copies only" and the repack "Move and add copies"

Changed repack CI test to test the Add Copies workflow (need reporting though)

Added a check for the repack just add copies workflow that verify that two archive routes does not point to the same tapepool

Remove the repack buffer if the repack expansion indicates that no Retrieve is necessary
Corrected the XrootDirectory::exist() method

This version contains several fixes required by the migration tools:
- fixes bug where quota bit was set when directory permissions have
  "executable by all" bit set
- inheritance of extended attributes
- better reporting of errors in case of failure

Removed BEGIN and END PL/SQL statements from OracleCatalogue.cpp.  To be put back if we experience performance problems.

Adding a non fatal ERROR for the missing sys.retrieve.req_id (for now) and a fatal one if we intercept some xrootd client failures during the past run

This reverts commit 17abf6f6.

Superseding of tape files by filesWrittenToTape() in now done in FSEQ ACS order in order to cover the corner case of the same tape copy being written more than once in the same archive batch