L2 performance bottlenecks found by the full-CLI harness (measured, not fixed)

Index: Known gaps

Status (re-verified 2026-10-09 at 456989f): NEGATIVE_RESULT. Parent section for measurement-only perf findings from scripts/check_l2_cli_perf.py (exists). Its subsections hold the actual status; this one only frames them.

Found while building scripts/check_l2_cli_perf.py. Recorded here rather than acted on: that work was explicitly measurement-only, and each of these is a production behaviour change needing its own design and its own evidence that the change is worth it. Every number below is a real local measurement (gcc 13.3.0 / castxml 0.7.0 / clang 18.1.3, 4 CPUs, Linux), reproducible with python scripts/check_l2_cli_perf.py --suite extended.