Consolidate all object storage configuration: only one bucket

This is a follow-up to #23345 (closed)

Today installing GitLab requires up to 13 different buckets, this adds a lot of friction to new installations as well as slowing down feature adoption.

Thanks to consolidated object storage configuration, developed in #23345 (closed) we now support one single credential for all the buckets.

The next step will be to use just one bucket.

added to epic &483

added Object Storage backend installation maintenancerefactor labels

mentioned in issue #23345 (closed)

marked this issue as related to #23345 (closed)

added automation:ml groupdistribution labels and removed Object Storage backend installation maintenancerefactor labels

This issue was automatically tagged with the label groupdistribution by TanukiStan, a machine learning classification model, with a probability of 0.8.

If this label is incorrect, please tag this issue with the correct group label as well as automation:ml wrong to help me learn from my mistakes.

added Object Storage backend installation maintenancerefactor labels

Setting label(s) devopsenablement sectionenablement based on groupdistribution.

added devopssystems sectioncore platform labels

mentioned in merge request !54244 (merged)

marked this issue as related to #300238 (closed)

mentioned in issue #300238 (closed)

mentioned in issue #326448

Requested by GitLab Premium customer as they've already migrated artifacts to a subdirectory in an existing bucket that's used for runner cache. GitLab team members can find out more in the ticket and SF.

@stanhu, @nolith - Hi! Is this groupdistribution ? I ask because it seems like there's back end changes needed, probably involving Workhorse, Fog and so on, in addition to support via configuration (such as Chef changes).

I can see the PM for groupdistribution prioritizing the latter, but not the former.

@bprescott_ As seen in gitlab-com/www-gitlab-com!71466 (merged), it's not clear which group should be owning this.

added customer label

Deployments created prior to the consolidated object store configuration are allowed to use 1 bucket with separate folders, and those are blocked to get migrated onto the consolidated configuration due to this limitation. Premium customer affected.

added WorkingGroupObjectStorage label

mentioned in issue gitlab-com/gl-infra/scalability#1683