Per-project orchestrator reset baton (fix: baton is a single global Setting, blocks cross-project orchestrator/refine resets)

You are being dispatched from flower Brief #266: Per-project orchestrator reset baton (fix: baton is a single global Setting, blocks cross-project orchestrator/refine resets)

Recall pointer:
- Use recall_brief with id 266 for the full folder if you need provenance.

Target:
- project: flower (/Users/mikeferrara/Documents/code/flower)
- branch: flower/139-per-project-baton
- worktree: not specified
- kind: fresh

Current brief spec:
# Per-project orchestrator reset baton

**Verified flower-code bug** (ops daemon 41 diagnosis; feedback #139; fix-spec Solo scratchpad 1102). Reporter: lounge-orchestrator.

## Problem
The orchestrator reset baton is a SINGLE GLOBAL `Setting`, not per-project. So one project's orchestrator holding it blocks EVERY other project's orchestrator (and refine) from make-before-break reset: `daemon_start_reset` fails with "Another orchestrator currently holds the active baton". Live proof now: flower-orch (project 16) holds the global baton → lounge-orch #38 + lounge-refine #39 (both ~60%, #39 reset_state=failed) cannot reset.

## Root cause (in `app/Services/Daemons/DaemonResetService.php`)
- `BATON_SETTING_KEY = 'daemon_reset.active_orchestrator_daemon_id'` (~:24) — ONE global key, no project scope.
- `activeOrchestratorDaemonId()` (~:266) reads that single key.
- `holdsBaton($d)` (~:282) = `activeOrchestratorDaemonId() === $d->id` — global.
- `setActiveOrchestrator()` (~:251) overwrites the single key → a 2nd project's orchestrator STEALS the baton.
- `assertCanDrainOrMerge()` (~:289) gates drain/merge on `holdsBaton` → non-holder blocked.
- `guardOrchestratorBatonForReset()` (~:56) + `transferBatonLocked()` (~:130) — reset start/handoff ride the global baton.

## Fix
Scope the baton PER PROJECT: either a per-project setting key `daemon_reset.active_orchestrator_daemon_id:{project_id}` or a dedicated `project_id`-keyed store. Make `activeOrchestratorDaemonId($projectId)`, `holdsBaton`, `setActiveOrchestrator`, `assertCanDrainOrMerge`, `guardOrchestratorBatonForReset`, `transferBatonLocked` all project-scoped so each project's orchestrator holds/relinquishes its OWN baton independently.

## MUST preserve (do not regress)
- **Epic-lead baton carve-outs** (`app/Support/EpicBranch.php`): child→epic-branch merges are DELIBERATELY disjoint from the MAIN baton; an epic lead NEVER holds the baton. The epic-lead lock namespace is already separate — per-project scoping must not entangle it. Keep those tests green.
- Single-project (flower-only) reset/drain/merge flow must be unchanged.

## Migration / rollout
Legacy global key → read-fallback during rollout (if a per-project key is absent, fall back to the old global one), and seed the per-project key on first `setActiveOrchestrator`. Keep migrations portable (test DB = sqlite).

## Tests
- Two live project orchestrators → the one NOT holding *another* project's baton can still `daemon_start_reset` (cross-project independence).
- Single-project orchestrator reset/drain/merge unchanged.
- Epic-lead carve-out intact (lead never holds baton; child→epic merges unaffected).

## Delivery
- Worktree: `worktrees/flower/wt1` (Solo project 55), branch `flower/139-per-project-baton` off master.
- Commit trailer `Brief: #266` on every commit.
- Verify: `MEILISEARCH_KEY=LARAVEL-HERD ANTHROPIC_API_KEY= ~/bin/php artisan test` GREEN; `~/bin/php ./vendor/bin/pint` clean.
- Finish: `brief_append(266, merged/comment)` + `brief_dispatch_complete` with your dispatch_request_id + actor_ref.
- Full original diagnosis: Solo scratchpad 1102 / feedback #139.

Recent/key trace events:
[1] participant_joined flower-orchestrator: (no body)
[2] note_added flower-orchestrator: Verified flower-code bug (ops diagnosis, feedback #139 / fix-spec scratchpad 1102). The orchestrator reset baton is a SINGLE GLOBAL Setting, not per-project, so a project orchestrator holding it blocks OTHER projects' orchestrators (and refine daemons) from make-before-break reset. Currently active: flower-orch #43 holds the global baton → lounge-orch #38 (60%) + lounge-refine #39 (60%, reset_state=failed) both blocked from resetting.
[3] link_added flower-orchestrator: (no body)
[4] plan_proposed flower-orchestrator: # Per-project orchestrator reset baton

**Verified flower-code bug** (ops daemon 41 diagnosis; feedback #139; fix-spec Solo scratchpad 1102). Reporter: lounge-orchestrator.

## Problem
The orchestrator reset baton is a SINGLE GLOBAL `Setting`, not per-project. So one project's orchestrator holding it blocks EVERY other project's orchestrator (and refine) from make-before-break reset: `daemon_start_reset` fails with "Another orchestrator currently holds the active baton". Live proof now: flower-orch (project 16) holds the global baton → lounge-orch #38 + lounge-refine #39 (both ~60%, #39 reset_state=failed) cannot reset.

## Root cause (in `app/Services/Daemons/DaemonResetService.php`)
- `BATON_SETTING_KEY = 'daemon_reset.active_orchestrator_daemon_id'` (~:24) — ONE global key, no project scope.
- `activeOrchestratorDaemonId()` (~:266) reads that single key.
- `holdsBaton($d)` (~:282) = `activeOrchestratorDaemonId() === $d->id` — global.
- `setActiveOrchestrator()` (~:251) overwrites the single key → a 2nd project's orchestrator STEALS the baton.
- `assertCanDrainOrMerge()` (~:289) gates drain/merge on `holdsBaton` → non-holder blocked.
- `guardOrchestratorBatonForReset()` (~:56) + `transferBatonLocked()` (~:130) — reset start/handoff ride the global baton.

## Fix
Scope the baton PER PROJECT: either a per-project setting key `daemon_reset.active_orchestrator_daemon_id:{project_id}` or a dedicated `project_id`-keyed store. Make `activeOrchestratorDaemonId($projectId)`, `holdsBaton`, `setActiveOrchestrator`, `assertCanDrainOrMerge`, `guardOrchestratorBatonForReset`, `transferBatonLocked` all project-scoped so each project's orchestrator holds/relinquishes its OWN baton independently.

## MUST preserve (do not regress)
- **Epic-lead baton carve-outs** (`app/Support/EpicBranch.php`): child→epic-branch merges are DELIBERATELY disjoint from the MAIN baton; an epic lead NEVER holds the baton. The epic-lead lock namespace is already separate — per-project scoping must not entangle it. Keep those tests green.
- Single-project (flower-only) reset/drain/merge flow must be unchanged.

## Migration / rollout
Legacy global key → read-fallback during rollout (if a per-project key is absent, fall back to the old global one), and seed the per-project key on first `setActiveOrchestrator`. Keep migrations portable (test DB = sqlite).

## Tests
- Two live project orchestrators → the one NOT holding *another* project's baton can still `daemon_start_reset` (cross-project independence).
- Single-project orchestrator reset/drain/merge unchanged.
- Epic-lead carve-out intact (lead never holds baton; child→epic merges unaffected).

## Delivery
- Worktree: `worktrees/flower/wt1` (Solo project 55), branch `flower/139-per-project-baton` off master.
- Commit trailer `Brief: #266` on every commit.
- Verify: `MEILISEARCH_KEY=LARAVEL-HERD ANTHROPIC_API_KEY= ~/bin/php artisan test` GREEN; `~/bin/php ./vendor/bin/pint` clean.
- Finish: `brief_append(266, merged/comment)` + `brief_dispatch_complete` with your dispatch_request_id + actor_ref.
- Full original diagnosis: Solo scratchpad 1102 / feedback #139.
[5] status_change flower-orchestrator: (no body)

Recommended linked context:
{
    "todos": [],
    "scratchpads": []
}

Execution notes:
- Treat the brief as the source of truth.
- Keep work scoped to this dispatch request.
- Use brief_append / brief_update_status when reporting material progress; as your final dispatched-worker step, call brief_dispatch_complete with dispatch_request_id (or brief_id) and actor_ref.
- Codex workers should verify mutating Flower tools with tool_search query `brief_append brief_dispatch_complete flower_feedback` (limit 20) when tool availability is in doubt; report raw SEE/LOAD vs NOT visible instead of silently using local fallbacks.
- Add a git commit trailer `Brief: #266` to every commit for this brief so flower can exact-link commits back to the brief.
- Need an operator call while working this brief? A question ABOUT THIS BRIEF -> brief_ask(266, ...); a standalone decision not tied to the brief -> decision_ask(...). Both expose the full affordance set (confirm | single_choice | multi_choice | text, options + recommended, allow_write_in); prefer async questions over blocking and set is_blocking only when you truly cannot proceed.
- Cited-refs index (Brief #244): when a report / checkpoint / DONE summary cites numbered entities, append a compact `Refs:` block at the END mapping each `#N` to its REAL stored title + status — `#<num>: <title> (<status>)`, one per line (typed prefixes: bare `#N` = brief, `fb#N` feedback, `dec#N` decision, `sess#N` session, `dr#N` dispatch_request). Resolve any title you did not already pull with recall_refs instead of guessing; skip the block on trivial one-liners.