Commenting this out from master due to broken dependency in CI

- CASTOR migration tools stay under CTA/migration
- upgrade tools are now under CTA/upgrade_db
- adds FindGRPC.cmake (required for migration tools)
- updates submodule to v0.50 (required for EOS gRPC API)

Moved disk-operations-related code into the new "disk" directory

Added the manually specified cmake variable named NoAcs.  Please note that this variable currently has no effect when building the CTA rpms

According to the documentation, this should limit the memory requirements of helgrind
from ~3GB to ~1GB per instance.
See http://valgrind.org/docs/manual/hg-manual.html#hg-manual.options

Adapted the in-RPM script generation.

The same variable is now used in all cases.

Added options, suppressions and re-implemented BackendVFS::AsyncLockfreeFetcher to try and make valgrind happy.

Now the submodule does not provide a CMakeLists.txt file, this is now
provided by the main project (eos_cta/CMakeLists.txt)

The cta-frontend and cta-taped rpms now create /var/log/cta

Increased queue depth in helgrind reports.
Added locking in a constructor to make helgrind happy.
Added suppression for helgrind finding races in synchronization functions (?)

Moved XrootCl exception and thrower to cta::exception