Resolve identities inside the scoped source aggregator

When a person lookup spans several systems, do not feed one canonical ID directly into every downstream source. Resolve the person once, then request only the needed work streams through the aggregator so each source can use its own verified identity mapping. Keep the stream list narrow to avoid unrelated sensitive data, and preserve per-source coverage and timestamps in the answer. If a direct source lookup disagrees with the aggregator, report the mismatch instead of treating either empty result as proof of absence.

Applies when: A current-work lookup combines code, issue-tracker, and coding-session sources. Keywords: identity mapping scoped aggregation source IDs coverage freshness work activity