fix(coding-agent): persisted vibe sessions across restarts

Vibe worker roster lived only in a process-local Map, so a resumed parent
session started with an empty registry and vibe_send failed with
"Unknown vibe session". Persist a versioned, parent-scoped lifecycle
journal (spawn/turn/tombstone events), rehydrate validated idle workers
through the persisted-subagent reviver on resume, and gate the flow with
generation/CAS protection so stale finalizers cannot clobber a
replacement worker. Killed transcripts stay readable but non-revivable;
mode-exit commits tombstones atomically with the mode change and rolls
back cleanly on storage failure.

Ported from @mastertyko's fork branch fix/vibe-session-persistence.

Fixes #5303

packages/coding-agent/src/collab/host.ts

@@ -614,10 +614,11 @@ export class CollabHost {
 			case "kill": {
 				const kill = async () => {
 					const ref = AgentRegistry.global().get(agentId);
-					if (ref && ref.status === "running" && ref.session) {
+					if (!ref) return;
+					if (ref.status === "running" && ref.session) {
 						await ref.session.abort({ reason: USER_INTERRUPT_LABEL });
 					}
-					await AgentLifecycleManager.global().release(agentId);
+					await AgentLifecycleManager.global().release(agentId, ref);
 				};
 				kill().catch(fail);
 				break;