refactor(bench): retire legacy Graph500 scale orchestration (#959)

[DecisionNerd]

Rebased onto current main (includes merged #1067 BenchExec Fly fixes). Retires duplicated legacy Graph500 scale orchestration after parity gate wiring.

#900 ladder evidence ingested

Fly S18/S19 qualification on ext4 volume (vol_40od6de197mn01p4, performance-16x):

Rung	Status	Wall time	Peak RSS
S18	passed	298s	7.0 GB
S19	passed	943s	14.0 GB

Commit 6255f9393362c996b5840566d255569761e881d7 image; S19 required 16 GB BenchExec memlimit.

make -C benchmarks parity-gate
# ready_for_retirement: true

Changes

• Remove legacy scripts/ci/validate-g500-ladder-qualification.py and related CI glue superseded by the isolated benchmarks/ harness.
• Rename .github/workflows/g500-certification.yml → progressive-ladder.yml and update gate registry.
• Extend parity gate criteria tracker; preserve historical migration fixtures.
• Ingest real S18/S19 rung bundles into fixtures/parity/ladder-bundle/.

Validation

make -C benchmarks smoke-python
make gate-registry-check
make -C benchmarks parity-gate  # ready_for_retirement: true
cd benchmarks && PYTHONPATH=harness uv run --locked python -m unittest tests.test_progressive_run tests.test_benchexec_authority -q

Fixes #959

  

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

Note

Retire legacy Graph500 scale orchestration and replace g500-certification with progressive-ladder

• Removes legacy G500 Make targets (bench-g500-scale20, bench-g500-ladder, g500-ladder-qualification, g500-ladder-qualification-check) and deletes associated CI scripts (build-g500-ladder-qualification.py, validate-g500-ladder-qualification.py) and their test suites.
• Renames the operator handoff workflow from g500-certification to progressive-ladder across progressive-ladder.yml, gate-registry.json, and test-ci-storage-policy.py. The registry now uses workflow id progressive-ladder with updated evidence contract, freshness, and sha semantics; operator_gates is cleared.
• Adds _legacy_orchestration_present helper to parity_gate.py that scans the Makefile, workflow file, and gate-registry.json for leftover g500-certification artifacts. The parity_gate_status function uses this to set the legacy_orchestration_retired_with_coverage criterion.
• Removes ignored Rust tests (run_ladder, scale20_public_facade_engineering_green, certification lifecycle tests) from scale_g500_ladder.rs and scale_g500_scale20.rs and updates module docs to point to the progressive-ladder operator.
• Adds new parity fixtures under benchmarks/fixtures/parity/ladder-bundle/ for S18/S19 rungs, manifest, and teardown inventory.
• Risk: Any out-of-tree consumers referencing the g500-certification workflow id or the removed Make targets will break; the operator_gates list in gate-registry.json is now empty, so consumers must use the workflows entry instead.

^{Macroscope summarized f2b623e.}

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Essentials

Run ID: 553c80a0-3333-4448-a148-2f891b947925

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

Warning

Billing warning: we have not been able to collect payment for this subscription for more than 72 hours. Please update the payment method or pay any pending invoices in Billing to avoid service interruption.

---

_{Comment @coderabbitai help to get the list of available commands.}