Storage family — first release

Added

Document	What it carries
`reference/storage/cloud-storage.md`	the five classes, their minimum durations, the four location types, and the bucket-default rule
`reference/storage/cloud-storage-access-control.md`	the five mechanisms, the uniform-access recommendation, and the 90-day one-way door
`reference/storage/persistent-disk-and-hyperdisk.md`	the three block families and the performance model that separates them
`reference/storage/filestore.md`	the current and legacy tiers, protocols, and the capacity floors
`reference/storage/errors.md`	the JSON API reason strings with cause, check and fix
`guides/storage/create-a-bucket-that-is-not-public.md`	six steps, ending in a verification that attempts what was forbidden
`guides/parity/storage.md`	S3 and Blob side by side, with the 365-against-180 archive row

The headline fact

ARCHIVE commits 365 days of storage charges, against 180 days for both AWS S3 Glacier Deep Archive and the Azure archive tier. Every lifecycle rule carried in from either cloud needs re-tuning, and nothing warns you — the charge simply appears.

Absences recorded rather than filled in

The Compute Engine disks overview does not address attaching a disk to multiple instances.
The Filestore service-tiers page says nothing about changing tier after creation.
The Cloud Storage classes page does not describe a rehydration step for ARCHIVE, where both AWS and

Azure document one for their deepest tiers. The parity guide records this as a difference and as an absence, because "no step is documented" is not the same claim as "no step exists".

Known missing

Object Lifecycle Management rule syntax in detail.
Bucket retention policies, object holds and Object Versioning.
Customer-managed encryption keys on buckets and disks (CMEK), beyond the KMS model in the identity family.
Snapshots and machine images.
Storage Transfer Service, which lives in the migration family instead.