feat: Refactor mesh server to use Arc<Node>, introduce JoinAcceptance for mesh joins, and enhance store listing with prefix filtering and deleted entry inclusion.

docs/architecture.md

@@ -59,14 +59,14 @@ Networking modes:
 - Identified by their Ed25519 public key.
 - Private key stored locally in `identity.key` (not replicated).
 - Node data stored in KV:
-  - `/nodes/{pubkey}/info` = static metadata (name, added_by, added_at)
-  - `/nodes/{pubkey}/status` = `active` | `dormant` | `disabled`
+  - `/nodes/{pubkey}/name` = display name
+  - `/nodes/{pubkey}/added_at` = timestamp when added
+  - `/nodes/{pubkey}/status` = `invited` | `active` | `dormant` (removal deletes keys)
   - `/nodes/{pubkey}/role` = `server` | `device` (optional, hints sync priority)
-  - `/nodes/{pubkey}/iroh` = Iroh NodeId for network connection
 - Peer invitation flow:
   1. Inviter runs `invite <peer_pubkey>` → writes `/nodes/{peer}/info` + `/status`
   2. Inviter shares their Iroh NodeId out-of-band (QR code, link, text)
-  3. Invited peer runs `connect <inviter_nodeid>` → syncs with inviter
+  3. Invited peer runs `join <inviter_nodeid>` → syncs with inviter
   4. Sync pulls `/nodes/{self}/info` + `/status` → peer is authorized
   5. `connect` implicitly adds inviter to peer's `/nodes/*` (mutual awareness)
 - Accepting = syncing. The invited peer discovers authorization by receiving the entries.
@@ -181,7 +181,7 @@ Each node stores logs as one file per author:
 Table              Key                     Value                      Purpose
 ─────────────────────────────────────────────────────────────────────────────
 kv                 Vec<u8> (key)           Vec<HeadInfo>              Current tips for each key
-applied_frontiers  [u8; 32] (author_id)    (u64 seq, [u8; 32] hash)   What's applied to this store
+AUTHOR_TABLE      [u8; 32] (author_id)    (u64 seq, [u8; 32] hash)   Per-author frontier tracking
 meta               Vec<u8>                 Vec<u8>                    Store metadata (incl. merkle_root)
 ```
 

docs/roadmap.md

@@ -127,10 +127,7 @@
 - [x] Entry ordering: Per-author streaming is correct (hash chain per author, HLC for cross-author).
 - [x] Multi-head sync fixed: SyncState now tracks HashSet of head hashes per author.
 - [x] Sync entry ordering: Entries sent in HLC order (merge-sort across authors) to ensure causal order.
-
-*Background Sync:*
-- [ ] Periodic sync with known peers
-- [ ] Track last sync time per peer
+- [x] `join_mesh` doesn't populate `node.root_store`: Fixed with `complete_join` method.
 
 ### Success Criteria
 
@@ -138,26 +135,43 @@
 - Works offline-first (sync when connected)
 
 **Post-M2 Refactoring:**
-- [ ] Unify `node.rs` from `lattice-cli` and `lattice-core`
-- [ ] Use prost for node status in store
+- [x] Unify `node.rs` from `lattice-cli` and `lattice-core`
+- [x] Move network code to `lattice-net`
 
 ---
 
 ## Milestone 3: Multi-Node Mesh
 
-**Goal:** N nodes form a gossip mesh with watermark consensus.
+**Goal:** N nodes form a gossip mesh for real-time sync.
 
 ### Deliverables
 
-- [ ] Gossip protocol
-- [ ] Watermark tracking & log pruning
-- [ ] Node invitation (sigchain membership)
-- [ ] Conflict detection (LWW resolution)
+**Phase 1: LatticeServer Refactor**
+- [ ] `LatticeServer` struct in `lattice-net` wrapping `Arc<Node>` + `Endpoint`
+- [ ] Move `join_mesh`, `sync_with_peer`, `sync_all` to `LatticeServer` methods
+- [ ] Encapsulate `spawn_accept_loop` inside `LatticeServer`
+- [ ] CLI uses `LatticeServer` instead of raw `Node` + `Endpoint`
+- [ ] Route sync command through `LatticeServer` (not raw functions)
+- [ ] Integration test: invite → join → sync end-to-end
+- [ ] Periodic background sync with known peers
+- [ ] Track last sync time per peer
+
+**Phase 2: Gossip Protocol**
+- [ ] Proto: `GossipAnnounce` message with author + latest seq + HLC
+- [ ] `LatticeServer::spawn_gossip_loop` for periodic announcements
+- [ ] On receiving announce: detect missing entries, trigger sync
+- [ ] Track last-seen per peer for staleness detection
 
 ---
 
 ## Future
 
+- remove_peer should be a transactional operation on store
+- Watermark tracking & log pruning
+  - Track minimum confirmed seq per author across all peers
+  - Log pruning: remove entries below watermark
+- Multi-KV-Store sync
+- Optimized sync on join. Only transfer current watermark state, then sync missing entries. This would allow pruning. Might need snapshot support in KV store.
 - Mobile (iOS/Android) clients
 - Key rotation
 - Secure storage (Keychain, TPM)