Relationship to the published artifacts

Fixtures target two artifact releases. The 1.0.x fixtures target release 1.0.1 (standards/ver/1.0.1/ver-record.schema.json) and carry ver_version "1.0"; the 1.1-draft fixtures target standards/ver/1.1-draft/ver-record.schema.json and carry ver_version "1.1".

Each release's flagship fixture tracks that release's published example, and a change to either example that this corpus does not follow surfaces as a fixture failure — which is the intent. valid/l3-full.json is byte-for-byte the published 1.0.1 example apart from the added signature. valid/l3-full-1-1.json is byte-for-byte standards/ver/1.1-draft/example-record-lineage.json with nothing added: the published 1.1 example carries no sign chain event, and §2.3.3 now forbids a signature without one, so adding a JWS to demonstrate a rule about signatures would be the exact defect class the 1.1 revision exists to remove. The two comparisons are two separate tests rather than one parametrised helper, because folding them together would mean weakening the stricter one.