Follow-up to #954, which added the same for the non-paging request already.

If pipelines have a large amount of jobs (> 100 at least), need to use
the paging API to get all jobs (as GitLab limits paging size to max. 100).
JobApi didn't support the include_retried flag yet, in the methods that
support paging (i.e. those that return Pager<Job>/Stream<Job>)

* #954 add include_retried for jobs api

* Apply suggestions from code review

* Fix compile error

---------

Co-authored-by: vladimir.molodkin <vladimir.molodkin@aliexpress.ru>
Co-authored-by: Jérémie Bresson <dev@jmini.fr>
Co-authored-by: Jeremie Bresson <jeremie.bresson@unblu.com>

Download a single artifact file from within the job's artifacts archive.
Only a single file is going to be extracted from the archive and streamed to a client.