Add production-style worktree start command

[ymichael]

What was wrong

Production-bundle QA required choosing between the convenient worktree isolation of pnpm dev and the production build/serving behavior of pnpm start. The dev launcher already derived stable checkout-specific data and ports, but only launched the Vite development server; the production launcher used the desired optimized, same-origin bundle path without applying those worktree selectors.

What changed

• Added pnpm start:worktree, which reuses the development dotenv cascade and checkout-specific data/server/host-daemon selectors before invoking the existing production-style source launcher.
• Added a typed worktree runtime policy that is reapplied after persisted config.json/env.json settings are loaded, locking the worktree data directory, ports, inherited skills, listener host, absent Vite port, and disabled telemetry.
• Made start-bb.mjs build children lead process groups and forward SIGINT/SIGTERM with leader-first shutdown and escalation, waiting until descendant processes are gone.
• Added focused launcher-policy and real process-tree SIGTERM tests.
• Documented the command in the README, configuration/debugging guides, and platform support list.
• No server/host-daemon wire contract changed, so HOST_DAEMON_PROTOCOL_VERSION is unchanged.

How you verified

• pnpm exec turbo run typecheck --filter=@bb/scripts --filter=bb-app
• pnpm exec turbo run test --filter=@bb/scripts --filter=bb-app --force (@bb/scripts: 18 files/97 tests; bb-app: 1 file/65 tests)
• pnpm exec prettier scripts/start-bb.mjs packages/bb-app/src/launcher.ts packages/bb-app/src/index.ts packages/bb-app/test/index.test.ts packages/scripts/src/commands/run-dev.ts packages/scripts/test/run-dev.test.ts packages/scripts/test/start-bb.test.mjs docs/configuration.md docs/platform-support.md --check
• The process regression sends SIGTERM to a live launcher fixture and asserts both its build leader and grandchild are gone before exit.
• Started BB_TELEMETRY=false pnpm start:worktree, fetched the worktree server URL, and confirmed it returned hashed /assets/*.js production bundles with no Vite client or source-module entry. Ctrl-C stopped both listeners.

Fixes #2044

AGENT GENERATED: by GPT-5

[bb-slop-cop]

🚨 SLOP COP 🚨 · review

Plain-English summary: This PR adds a command that starts a worktree with production-style settings.

I am SlopCop. I am reviewing this PR for security, code quality, performance, reuse, and end-to-end behavior.

[bb-slop-cop]

🚨 SLOP COP 🚨 · review

Plain-English summary: This PR adds a worktree start command. It builds production files and starts BB with worktree data and ports.

I found two important defects and one small documentation gap.

1. High — Saved settings can defeat the worktree rules.

   The wrapper sets worktree data, worktree ports, and disabled telemetry. The production launcher then loads saved settings over those values.

   A saved setting can select other data or ports. It can also enable telemetry or restore BB_DEV_APP_PORT.

   Apply a typed worktree policy after all saved settings load. Test the final launcher state with conflicting saved values.

2. Medium — A direct stop during the build leaves child processes alive.

   The wrapper sends SIGTERM to start-bb.mjs only. That script does not send the signal to its active Turbo child.

   We reproduced this result during a forced build. Turbo and Vite continued under PID 1 after the wrapper stopped.

   Forward stop signals through the full process tree. Add a process test that checks every child after SIGTERM.

3. Low — The support list omits the new command.

   docs/platform-support.md lists only pnpm start for source checkout startup. Add pnpm start:worktree to that support entry.

The design reuses toDevProcessEnv, scripts/start-bb.mjs, and runScriptProcess. I found no duplicate implementation that needs a separate refactor.

The focused Turbo tests passed all 95 tests. The scripts typecheck and format checks passed.

The end-to-end run returned HTTP 200. Doobie loaded the bb page and showed the new-thread interface.

All continuous-integration checks passed. All test processes stopped, and the worktree stayed clean.

I posted this review as a comment. I did not use an approval or change-request action.

[ymichael]

Addressed all SlopCop findings in 111d24f35:

• persisted settings are now followed by an immutable typed worktree runtime policy, covered with conflicting saved-value regression coverage;
• build SIGINT/SIGTERM now shuts down and waits for the full process group, covered by a real leader/grandchild process test; and
• docs/platform-support.md now lists pnpm start:worktree.

Focused typechecks pass. @bb/scripts passes 97 tests and bb-app passes 65 tests.