fix: use transcript usage as fallback for /status token display

When using custom providers like LM Studio, Ollama, or DashScope,
token counts in /status show as 0 because the agent meta store
does not always have usage data populated for these providers.

Fix: set includeTranscriptUsage: true in both /status command and
the session_status tool. This enables the existing fallback path
that reads usage from the session transcript JSONL file when the
meta store has zero/missing token counts.

The merge logic already guards against overwriting valid data:
- totalTokens: only updated when zero or transcript value is larger
- inputTokens/outputTokens: only filled when zero/missing
- model/contextTokens: only filled when missing

Fixes #54995

src/agents/tools/session-status-tool.ts

@@ -499,6 +499,7 @@ export function createSessionStatusTool(opts?: {
         skipDefaultTaskLookup: true,
         primaryModelLabelOverride: primaryModelLabel,
         ...(providerForCard ? {} : { modelAuthOverride: undefined }),
+        includeTranscriptUsage: true,
       });
       const fullStatusText =
         taskLine && !statusText.includes(taskLine) ? `${statusText}\n${taskLine}` : statusText;

src/auto-reply/reply/commands-status.ts

@@ -345,7 +345,7 @@ export async function buildStatusText(params: {
     subagentsLine,
     taskLine,
     mediaDecisions: params.mediaDecisions,
-    includeTranscriptUsage: false,
+    includeTranscriptUsage: true,
   });
 
   return statusText;