feat(history): 会话消息持久化 + 主/次级会话（design/16）

- runtime/history: Store + HistoryStore 接口（Append/List 游标分页/Recent/DeleteSession/NewSubSession/ListSessions），主会话固定 web、次级 web:<uuid>
- db 迁移 0003: chat_sessions + chat_messages + idx_chat_msgs_session_created（事务幂等）
- gateway: GET/POST /sessions、GET /sessions/{id}/messages 热分页、DELETE /sessions/{id}（主不可删 403）；Deps.History 可空注入
- 编排层落库: session_persist.go（submitInternal 写用户消息、completeTurn 写助手消息，拒绝的消息不落库）
- WS: wsEvent 加 sessionId 定向 + history 回放帧（握手上行后回放主会话最新 50 条，thinking 剥离）
- wire.go 装配 History 到 gateway.Deps.History
- 端到端实测: POST /sessions 建次级 201 / GET /sessions 列表 / POST /chat 后 /messages 读到 user+assistant / DELETE 次级成功 / DELETE 主会话 403

internal/services/gateway/server.go

@@ -37,6 +37,7 @@ import (
 	"git.yeij.top/AskaEth/Cyrene_v2/internal/logger"
 	"git.yeij.top/AskaEth/Cyrene_v2/internal/persona/loader"
 	personareg "git.yeij.top/AskaEth/Cyrene_v2/internal/persona/registry"
+	"git.yeij.top/AskaEth/Cyrene_v2/internal/runtime/history"
 	"git.yeij.top/AskaEth/Cyrene_v2/internal/runtime/knowledge"
 	"git.yeij.top/AskaEth/Cyrene_v2/internal/runtime/loop"
 	"git.yeij.top/AskaEth/Cyrene_v2/internal/runtime/memory"
@@ -102,6 +103,19 @@ type MemoryStore interface {
 	List(ctx context.Context) ([]memory.MemoryItem, error)
 }
 
+// HistoryStore is the persistent chat-history store (design/16). *runtime/history.Store
+// is the production implementation. When nil the session/message endpoints degrade: list
+// sessions falls back to the channel-session view, and the create/messages/delete
+// endpoints answer with a clear 501.
+type HistoryStore interface {
+	Append(ctx context.Context, msg history.Message) error
+	List(ctx context.Context, sessionID string, before int64, limit int) ([]history.Message, error)
+	Recent(ctx context.Context, sessionID string, n int) ([]history.Message, error)
+	DeleteSession(ctx context.Context, sessionID string) error
+	NewSubSession(ctx context.Context, ownerID string) (string, error)
+	ListSessions(ctx context.Context) ([]history.SessionMeta, error)
+}
+
 // IoTStore lists IoT devices. *iot.Manager is the production implementation.
 type IoTStore interface {
 	List(ctx context.Context) ([]iot.Device, error)
@@ -179,6 +193,9 @@ type Deps struct {
 	Personas PersonaStore
 	// Memories lists memory entries. Optional.
 	Memories MemoryStore
+	// History is the persistent chat-history store (design/16). Optional; when nil the
+	// session/message endpoints degrade (list falls back to channel sessions, others 501).
+	History HistoryStore
 	// IoT lists IoT devices. Optional.
 	IoT IoTStore
 	// Scheduler audits sub-agent context for the subagent_ask WS route.
@@ -229,6 +246,7 @@ type Server struct {
 	sessions    SessionStore
 	personas    PersonaStore
 	memories    MemoryStore
+	history     HistoryStore
 	iot         IoTStore
 	auditor     SubagentAuditor
 	taskOutputs TaskOutputStore
@@ -270,6 +288,7 @@ func New(cfg *config.Config, deps Deps) (*Server, error) {
 		sessions:    deps.Session,
 		personas:    deps.Personas,
 		memories:    deps.Memories,
+		history:     deps.History,
 		iot:         deps.IoT,
 		auditor:     deps.Scheduler,
 		taskOutputs: deps.Tasks,
@@ -305,6 +324,15 @@ func (s *Server) Handler() http.Handler {
 	mux.Handle("GET /api/v1/sessions", readAuth(http.HandlerFunc(s.handleListSessions)))
 	mux.Handle("GET /api/v1/personas", readAuth(http.HandlerFunc(s.handleListPersonas)))
 
+	// Session/model reads and writes are the owner's private history; owner/admin only
+	// (chat uses the same roles). Session metadata in GET /sessions stays readAuth for
+	// any authenticated role, but the message content (and thinking) is owner/admin-only
+	// per design/16 §6. Creating and deleting a sub session mutate persistent state.
+	historyAuth := s.auth.Authorize(auth.RoleOwner, auth.RoleAdmin)
+	mux.Handle("GET /api/v1/sessions/{id}/messages", historyAuth(http.HandlerFunc(s.handleListMessages)))
+	mux.Handle("POST /api/v1/sessions", historyAuth(http.HandlerFunc(s.handleCreateSubSession)))
+	mux.Handle("DELETE /api/v1/sessions/{id}", historyAuth(http.HandlerFunc(s.handleDeleteSession)))
+
 	// Realtime.
 	mux.HandleFunc("/ws/chat", s.handleWS)
 
@@ -425,9 +453,10 @@ func newOrchestrator(s *Server) *Orchestrator {
 			return s.hub.runTurn(ctx, sessionID, messages, env)
 		},
 		Emit: func(sessionID, typ string, data map[string]any) {
-			s.hub.broadcast(wsEvent{Type: typ, Data: data})
+			s.hub.broadcast(wsEvent{Type: typ, SessionID: sessionID, Data: data})
 		},
-		Log: s.log,
+		Log:     s.log,
+		History: s.history,
 	})
 }