fix(turn-metrics): exact per-turn placement via turnId, decode-only TPS, step-level retention

Placement
- Chunks carry their producing turn: StoredChunk.turnId (backend [email protected]) for
  committed rows (persisted in the IDB cache; old cached rows are upgraded on
  sync) and the live event's turnId for provisional ones (the optimistic echo is
  tagged on turn-start).
- interleaveTurnMetrics attributes every group to its turn (turnId, else the
  turn owning its stepId) and puts the turn total after the turn's LAST visible
  group. No user-message counting: sending, streaming, steering, images, errored
  turns and chat-limit unloading can no longer shift metrics between turns — at
  any point during a turn, not just after it ends. A turn cut in half by the
  chat limit keeps its total under its visible tail.
- Legacy rows (no turnId) fall back to an order-preserving, count-guarded match
  that leaves a turn blank rather than show another turn's numbers.
- A turn that errors re-syncs metrics (the backend now persists them).

TPS
- Step TPS = outputTokens / decodeMs only (no genTotalMs fallback — it includes
  time-to-first-token). Turn TPS sums tokens and decode time over timed steps
  only. Tool execution is excluded by the backend.

Display
- Retention = this turn's first-step cache reads / previous turn's last-step
  prompt (step-level; turn sums made it pin at 100%).
- Turn row shows output tokens + input processed (with cached), instead of a
  misleading 'total tok' that summed every step's full prompt.

1 files changed, 9 insertions, 1 deletions

diff --git a/src/core/chunks/groups.ts b/src/core/chunks/groups.ts
index c1c80c9..9df0fa5 100644
--- a/src/core/chunks/groups.ts
+++ b/src/core/chunks/groups.ts
@@ -22,6 +22,8 @@ export type RenderGroup =
       readonly stepId: string;
       readonly entries: readonly ToolBatchEntry[];
       readonly provisional: boolean;
+      /** The producing turn (from the batch's first call chunk), when known. */
+      readonly turnId?: string;
     };
 
 /**
@@ -79,6 +81,7 @@ export function groupRenderedChunks(rendered: readonly RenderedChunk[]): readonl
       // (Svelte `each_key_duplicate` aborts the reactive flush — tab appears dead).
       const seenCallIds = new Set<string>();
       let provisional = false;
+      let turnId: string | undefined;
       for (const inner of rendered) {
         if (inner.chunk.type === "tool-call" && inner.chunk.stepId === stepId) {
           if (seenCallIds.has(inner.chunk.toolCallId)) continue;
@@ -86,9 +89,14 @@ export function groupRenderedChunks(rendered: readonly RenderedChunk[]): readonl
           const result = resultByCallId.get(inner.chunk.toolCallId) ?? null;
           entries.push({ call: inner.chunk, result });
           if (inner.provisional) provisional = true;
+          turnId ??= inner.turnId;
         }
       }
-      groups.push({ kind: "tool-batch", stepId, entries, provisional });
+      groups.push(
+        turnId !== undefined
+          ? { kind: "tool-batch", stepId, entries, provisional, turnId }
+          : { kind: "tool-batch", stepId, entries, provisional },
+      );
       continue;
     }