feat: persist per-tab token/cache usage across reload

Persist usage as invisible type:"usage" chunk rows (side channel):
- core: add "usage" ChunkType + UsageData; exclude usage rows from
  getChunksForTab/getTotalChunkCount; add getUsageStatsForTab aggregate
  (exported from barrel); defensive skip in groupRowsToMessages.
- api: agent-manager accumulates per-attempt usageRows and flushes them in
  the same atomic appendChunks call as the turn's content (discarded on a
  superseded fallback attempt). GET /tabs enriches rows with usageStats.
- frontend: hydrateFromBackend seeds cacheStats from usageStats (reload only;
  no re-seed on statuses reconnect, so no double-count with live events).

Tests: core DB-backed usage persistence/aggregate; api usage-row-per-event +
fallback discard; routes GET /tabs usageStats; frontend hydrate seed +
no-double-count + live-accumulation-after-seed. 495 -> 509 passing.

1 files changed, 7 insertions, 0 deletions

diff --git a/packages/frontend/src/lib/tabs.svelte.ts b/packages/frontend/src/lib/tabs.svelte.ts
index 3fd7e5f..54f17d9 100644
--- a/packages/frontend/src/lib/tabs.svelte.ts
+++ b/packages/frontend/src/lib/tabs.svelte.ts
@@ -755,6 +755,12 @@ export function createTabStore() {
 			keyId?: string | null;
 			modelId?: string | null;
 			parentTabId?: string | null;
+			// Backend usage aggregate (GET /tabs). Structurally identical to
+			// CacheStats, so it seeds `cacheStats` directly on reload. Seeding
+			// happens ONLY here (hydrate runs when tabs.length === 0, i.e. a true
+			// reload) — never on `statuses` reconnect or `turn-sealed` — so the
+			// persisted aggregate and in-session live `usage` events never overlap.
+			usageStats?: CacheStats | null;
 		}> = [];
 		try {
 			const res = await fetch(`${config.apiBase}/tabs`);
@@ -843,6 +849,7 @@ export function createTabStore() {
 				chunkLimit: appSettings.chunkLimit,
 				oldestLoadedSeq: win.oldestSeq,
 				totalChunks: win.total,
+				cacheStats: row.usageStats ?? undefined,
 			};
 		});