Align metrics to DCAT-US 3.0 data model

### User story

As the Data.gov platform team, we need to ensure that stored and surfaced metrics align with the DCAT-US 3.0 data model so that metrics remain accurate and useful as the catalog evolves.

### Background

DCAT-US 3.0 introduces distinctions between catalog-level and dataset-level records. Metrics keyed to dataset identifiers need to remain stable and correctly mapped as the catalog transitions. This ticket covers the analysis and implementation work to confirm that metric identifiers, dimensions, and display logic are aligned with the DCAT-US 3.0 model before the metrics system is built out.

### Acceptance criteria

- Given the DCAT-US 3.0 data model, when metrics are stored, then they are keyed to the dataset-level identifier consistent with DCAT-US 3.0 definitions.
- Given a dataset record changes identifier during migration, when metrics are retrieved, then the system handles identifier continuity or flags the break clearly.
- Given metrics are displayed on the organization page, when a dataset is listed, then the metric shown corresponds to the correct DCAT-US 3.0 dataset record.

### Open questions

- How does DCAT-US 3.0 distinguish between catalog and dataset for identifier purposes in our implementation? (DCAT standards lead to advise.)
- What happens to metrics continuity for datasets that went through DCAT-US 3.0 migration with identifier changes?

### Dependencies

- DCAT-US 3.0 migration (September 2026 compliance deadline).
- DCAT standards lead and back-end developer to advise on identifier model.

No files, tests, or entry points are named. Start by reviewing the DCAT-US 3.0 data model and the planned migration, then clarify the dataset-level identifier, migration continuity behavior, and organization-page display requirements with the DCAT standards lead and back-end developer. Done means all three acceptance criteria are verified.

Written by the indexing model from the issue text.