fix(test): isolate soak harness daemon runtime

[astandrik]

What does this PR do?

Part of #1696 (audit ledger), follow-up to #1691/#1695.

scripts/soak-test.sh claimed to "isolate daemon coordination from interactive CBM
sessions" through a private CBM_CACHE_DIR, but only CBM_RUNTIME_DIR moves the
daemon rendezvous. The soak must be the only client of the daemon it measures (it
asserts that a session crash and the final shutdown each stop the daemon), so on a
shared rendezvous it either stopped the operator's daemon or was refused with a
cache-root conflict.

• scripts/soak-test.sh sources scripts/test-runtime.sh and calls
  cbm_test_runtime_init; off native Windows the helper's owner-only cache is the
  soak cache, so the daemon log path is unchanged in shape. soak_cleanup now runs
  cbm_test_runtime_cleanup "$BINARY" instead of rm -rf on the cache: the private
  daemon is stopped first and the root is kept when it will not stop. The
  native-Windows stamped root and its icacls handling are untouched
  (tests/test_soak_daemon_recovery_contract.sh pins them). Two FAIL messages no
  longer say "shared daemon".
• tests/test_soak_runtime_isolation_contract.sh: environment-probe fixture,
  same assertions as the smoke contract, RESULTS_DIR kept out of the checkout.
  Fails on main with FAIL: soak-test exposed the caller CBM_RUNTIME_DIR to a product process; passes with this change. Wired as Step 0d2 in scripts/test.sh.

scripts/soak-legs.sh starts no product process and needs no change.

Checklist

• Every commit is signed off (git commit -s)
• Tests pass locally (make -f Makefile.cbm test) — shell-only change; the new
  contract plus test_soak_daemon_recovery_contract, test_venue_parity_contract
  and test_runtime_isolation_contract were run; the full C suite was not run
  on this host
• Lint passes (make -f Makefile.cbm lint-ci) — no C changes
• New behavior is covered by a test (reproduce-first for bug fixes)

[github-actions]

Thanks for opening this — it has been seen, and it is queued.

This note is automated, but it is not a brush-off: it exists so you know where your PR stands instead of having to guess from silence.

Current review status: working through a backlog. 0.9.1-rc.1 is out, so the release freeze that held reviews is over — but it left a large queue of open pull requests behind it, and we are reading through them oldest-first. The background is in discussion #1144.

What that means for this PR, concretely:

• It will not be closed for inactivity. No stale bot touches pull requests here.
• It may still sit a while before a human reads it. That is on us, not on you.
• Older PRs are read first, so a recent one is not being skipped — it is behind a queue.

Things that will genuinely speed it up whenever review does happen:

• Keep it rebased on main — the tree is moving quickly right now, and a conflicting branch cannot be reviewed as the diff you intended.
• Get CI green, or say which failures you believe are pre-existing.
• Keep the change to one claim. Bundled features and refactors get split before they get merged, which costs you a round trip.
• Every commit needs a sign-off (git commit -s) — CI enforces DCO.

If this fixes a bug, a reproduction we can run is worth more than a description of the symptom.

Thanks for contributing, and sorry in advance for the wait.

[Copilot]

🟡 Changes recommended

Address the native-Windows cleanup and setup-failure root-leak issues.

Get a fresh assessment by requesting another Copilot review.

Pull request overview

This PR isolates the soak-test daemon runtime and cache from interactive sessions and adds regression coverage.

Changes:

• Uses private runtime and cache directories.
• Adds a runtime-isolation contract test.
• Wires the contract into scripts/test.sh.

File summaries

File	Summary
tests/test_soak_runtime_isolation_contract.sh	Verifies caller runtime and cache paths are not exposed.
scripts/test.sh	Runs the new isolation contract.
scripts/soak-test.sh	Initializes and cleans up the private runtime; two moderate cleanup issues remain on native Windows.

Review details

• Files reviewed: 3/3 changed files
• Comments generated: 2
• Review effort level: Lite

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.