Compact runner reports and refresh debt counts

[Sapientropic]

Summary

• Make python tools/aippocampus/run_tests.py --report-json compact by default, with the full tier catalog behind --detail full.
• Refresh architecture debt register current_count rows after reviewing positive/negative drift against real headroom; no budgets were raised.
• Broaden the same warning-shape fix to debt compact reports: warning-bearing but nonfatal debt output now says status=advisory_action_recommended instead of clean pass.

Closes #2964.
Closes #2965.

Closeout

Closeout class: complete
Evidence level: behavior_run
remaining_gap: none for #2964/#2965. The pre-existing single-digit guard-pressure advisory remains parked under the existing architecture-debt policy and is not part of this closeout.

Debt removed / recurrence guard:

• Make run_tests --report-json compact by default with full tier catalog behind detail #2964 removed the default operator-catalog dump from the no-run runner report; catalog inventories now stay in explicit full detail.
• Refresh architecture debt register current-count drift on main #2965 removed 12 stale architecture-register current_count rows. The positive drift rows were reviewed before refresh, and all retained non-budget-raised headroom.
• Same-family projection issue found during implementation: compact debt reports could show warning_count > 0 while status=pass; this PR changes that to advisory status without hard-failing historical warning debt.

Before / after inventory:

• Make run_tests --report-json compact by default with full tier catalog behind detail #2964 before on main: python tools/aippocampus/run_tests.py --report-json emitted 74284 bytes and exposed top-level tier_definitions, tier_aliases, tier_shrink_replacement_lanes, timing_artifacts, and known_limitations.
• Make run_tests --report-json compact by default with full tier catalog behind detail #2964 after compact: python tools/aippocampus/run_tests.py --report-json emitted 3764 bytes, kind=aippocampus_test_tier_report_compact, status=advisory_action_recommended, warning_count=4, and no full catalog keys.
• Make run_tests --report-json compact by default with full tier catalog behind detail #2964 after detail: python tools/aippocampus/run_tests.py --report-json --detail full emitted 74284 bytes, kind=aippocampus_test_tier_report, with full catalog keys preserved.
• Refresh architecture debt register current-count drift on main #2965 before: python tools/aippocampus/docs/debt_report.py --headroom-only --json --detail full reported 12 count drifts: harmless_small_drift=6, positive_drift=5, negative_drift=1.
• Refresh architecture debt register current-count drift on main #2965 review before refresh: positive rows still had headroom after true counts: hooks/claude_code.py margin 30, recall/feedback/events.py margin 45, recall/semantic_cue_cache.py margin 26, tests/aippocampus/test_update_sync.py margin 41, instruction_surface_owner_registry.py margin 19. The negative row update/plugin_installer.py increased margin to 67. No budgets were raised.
• Refresh architecture debt register current-count drift on main #2965 after refresh: python tools/aippocampus/docs/debt_report.py --headroom-only --json --detail full reported count_drift=0 and only the pre-existing architecture_debt_single_digit_guard_pressure advisory warning.
• Refresh architecture debt register current-count drift on main #2965 after compact: python tools/aippocampus/docs/debt_report.py --headroom-only --json emitted 511 bytes, kind=aippocampus_debt_headroom_compact, ok=true, status=advisory_action_recommended, warning_count=1, and no refresh command because count drift is gone.

Tests added / rewritten:

• Added compact vs full run_tests --report-json projection coverage and a warning-action regression in tests/aippocampus/test_run_tests_tiers.py.
• Added debt compact warning-status coverage in tests/aippocampus/test_debt_report.py.
• No bad tests were deleted; the changed tests now protect compact/detail boundaries and advisory status semantics rather than field presence alone.

Known remaining debt:

• The debt drift mechanism remains advisory, not a hard gate. It already has a changed-surface/preflight reminder path, and this PR makes advisory warnings impossible to read as clean pass in compact debt output.
• The remaining architecture_debt_single_digit_guard_pressure warning is pre-existing, parked/freeze-except-delete-or-split debt and is not closed by Refresh architecture debt register current-count drift on main #2965.

Verification

• python -m unittest tests.aippocampus.test_debt_report tests.aippocampus.test_run_tests_tiers -v -> 58 tests OK.
• python -m unittest tests.aippocampus.test_architecture_boundaries tests.aippocampus.test_docs_health -v -> 108 tests OK.
• python tools/aippocampus/docs/check_docs_health.py -> OK, with two existing IA warnings.
• python -m ruff check tools/aippocampus/run_tests.py tools/aippocampus/docs/debt_report_projection.py tests/aippocampus/test_run_tests_tiers.py tests/aippocampus/test_debt_report.py -> passed.
• python tools/aippocampus/changed_surface_preflight.py --json -> pass.
• python tools/aippocampus/changed_surface_preflight.py --json --detail full --mode closeout --base origin/main -> runnable gates passed; PR-body audit was left for after PR creation, as intended.