Chat-wake tasks live and die with their reply run

Of the 202 tasks stranded in the backlog on July 9, 200 were "Reply to ..."
wake vehicles: every agent @mention creates a task plus a queued run, the
reply goes out, and nothing ever closed the task — a task with a completed
run is never enqueued again, so each one became a permanent zombie row. On
top of that, every message in an active thread minted a fresh wake even when
the target already had one queued, multiplying runs that would each read the
same conversation.

The lifecycle contract, made explicit:
- The reply run IS the wake task's deliverable. Completing a chat_mention
  run now closes its task (backlog/in_progress -> done, via the audited
  status path). Tasks an agent moved to review are left alone.
- One pending wake per (agent, thread): wakeFromMention returns the queued
  run instead of minting a duplicate — that run reads the thread's newest
  messages when it executes anyway.
- Migration 054 heals existing rows: wake tasks whose runs are all spent
  close as done; in-flight wakes keep working. This also stops Fix 1's
  retry-budget healing from resurrecting 31 stale replies on next launch.

The chat-governance loop test now completes each pending wake run between
posts, matching how the exchange actually interleaves in production.

src/main/services/chat-agent-wake-service.ts

@@ -37,6 +37,11 @@ export class ChatAgentWakeService {
 
   wakeFromMention(input: SendMessageInput, agentId: string, messageId: string, threadId: string, now: string): AgentWakeResult {
     const agent = this.requireRunnableAgent(input.orgId, agentId, "Agent mention");
+    // One pending wake per (agent, thread): the queued run reads the thread's
+    // newest messages when it executes, so waking again before it starts would
+    // spend a whole run to answer the same conversation twice.
+    const pending = this.pendingWakeRun(input.orgId, agentId, threadId);
+    if (pending) return { targetId: agentId, taskId: pending.taskId, runId: pending.runId };
     const workspace = this.workspaceForAgent(input.orgId, agent.workspace_id);
     const from = this.principalName(input.orgId, input.senderType, input.senderId);
     const space = this.spaceRow(input.orgId, input.spaceId);
@@ -81,6 +86,16 @@ export class ChatAgentWakeService {
     return { targetId: agent.id, taskId, runId };
   }
 
+  private pendingWakeRun(orgId: string, agentId: string, threadId: string): { taskId: string; runId: string } | null {
+    const row = this.db.prepare(`
+      select id, task_id from task_runs
+      where org_id = ? and agent_id = ? and state = 'queued' and trigger = 'chat_mention'
+        and json_extract(secret_scope_json, '$.sourceThreadId') = ?
+      limit 1
+    `).get(orgId, agentId, threadId) as { id: string; task_id: string } | undefined;
+    return row ? { taskId: row.task_id, runId: row.id } : null;
+  }
+
   private requireAgentMention(mentions: ResolvedMention[]): string {
     const target = mentions.find((mention) => mention.target.type === "agent")?.target;
     if (!target || target.type !== "agent") throw new Error("Slash command /agent requires an agent mention.");