Add benchmark measurement coverage audit

[yvette-carlisle]

Summary

• adds a fresh benchmark measurement coverage audit for ELF memory-system competitiveness
• records current fixture aggregate and fresh ELF/qmd live real-world adapter counters
• separates measured claims from non-claims across all 17 tracked projects and scenario strengths
• adds a machine-readable JSON summary for later report generation

Scope

Report-only. This does not implement ELF optimizations or queue implementation work. It records what is actually measured and which measurement reports should guide future iteration.

Verification

• cargo make real-world-memory
• cargo make real-world-memory-live-adapters
• jq empty docs/research/2026-06-11-measurement-coverage-audit.json
• git diff --check
• project coverage grep for all 17 tracked projects in the new report