data_validation: point riskassessment central schema at the dataset-local one

[Prasad-desh]

Summary

Refs #76. Follows the pattern from #73. Supports the Data Security Initiative's data_validation workstream.

Converts data_validation/schemas/riskassessment.schema.json into a $ref pointer to datasets/riskassessment_dataset/schema.json, which is authoritative per the policy ratified in #62.

Reference check: grep -rn "riskassessment.schema" returns no matches. Nothing validates against the central file; only READMEs link to the dataset folder.

Evidence:

• datasets/riskassessment_dataset/validate.py: all 23 entries pass the local schema
• Validating the same 23 entries against the central draft-07 schema: 23/23 fail (missing deployment_scenario, components_assessed, dsgai_risks_identified; assessment_id does not match ^RA-\d{4}$)

Divergences recorded in $comment:

• Draft version: draft-07 to 2020-12, with additionalProperties: false
• assessment_id: RA-NNNN to DSGAI-RA- slug
• components_assessed (string list) to typed components with role enum and sensitivity tier
• dsgai_risks_identified with single risk_rating to risks_identified with inherent and residual ratings plus required rationale
• existing_controls, control_gaps, framework_alignment from free text to structured arrays
• industry (free text) to industry_sector enum
• New required fields: scenario_name, entry_type, deployment_pattern, description, date_added

Analysis drafted with AI assistance; I reviewed the diff and verified results by running the validators locally.

DSGAI mapping

None. Schema housekeeping only; no dataset entries are added or changed.

Checklist

• Data is anonymized (no PII, credentials, or org names): N/A, no data in this change
• Entries reference at least one DSGAI ID where applicable: N/A, no entries changed
• Validation scripts pass (python run_all_checks.py): run_all_checks.py is currently a stub ("Validators not yet implemented"). Ran datasets/riskassessment_dataset/validate.py instead: 23/23 entries pass.
• One entry per file for dataset contributions: N/A, not a dataset contribution

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[Prasad-desh]

@emmanuelgjr — opened this per the plan above. grep -rn "riskassessment.schema" confirmed nothing references the central file, and the concrete divergences are recorded in the $comment.

One note beyond this PR: the checklist's python run_all_checks.py is currently a stub ("Validators not yet implemented"), so I validated with datasets/riskassessment_dataset/validate.py instead — 23/23 entries pass. Happy to open a separate issue for that if it's useful.

I'll take vulnerability.schema.json next if no one else has claimed it.

[emmanuelgjr]

Verified locally: all 23 entries pass the dataset-local schema and the new $ref pointer, and all 23 fail the old central file. Nothing references the central file. Same pattern as #73. Thanks!