Fixed typo in run_pipeline! function
  -> Open3.pieline to Open3.pipeline

Adjusted run_pipeline! function
  to match rubocop syntax requirements
Adjusted run_pipeline! function
  to use variable instead of string interpolation

Changed run_pipeline! function to have use $ in regex
Changed run_pipeline! function to use Open3.pipeline
instead of Open3.pipeline_start

Changed the run_pipeline! function to instead use Open3.pipeline_start
In doing so, the code is able to grab the last STDERR. In the case
where the error is the one from older tar versions (relating to ./),
it will not raise a Backup::Error. All other instances of command
failure will result in Backup::Error being raised.

Specifically, the `ListAllMyBuckets` permission.

This works if you know the directory exists.

See more:

* https://gitlab.com/gitlab-org/gitlab-ce/merge_requests/23981
* https://stackoverflow.com/a/12288581/1992201

Pool repositories are persisted in the database, and when the DB is
restored, the data need to be restored on disk. This is done by
resetting the state machine and rescheduling the object pool creation.

This is not an exact replica of the state like at the time of the
creation of the backup. However, the data is consistent again.

Dumping isn't required as internally GitLab uses git bundles which
bundle all refs and include all objects in the bundle that they require,
reduplicating as more repositories get backed up. This does require more
data to be stored.

Fixes https://gitlab.com/gitlab-org/gitaly/issues/1355

Partially addresses #47424.

Had to make changes to spec files because
stubbing methods on frozen objects is a mess
in RSpec and leads to failures:

https://github.com/rspec/rspec-mocks/issues/1190