fix: show all batch log lines in Live Output (was truncated to last 30)

Frontend: removed .slice(-30) — now shows all lines from backend.
Backend: bumped log line limit from 200 to 2000 so full Forge game
output is captured for debugging.

routes/lab.py

@@ -454,7 +454,7 @@ async def get_log(batchId: str):
     state = active_batches.get(batchId)
     if not state:
         raise HTTPException(404, f"Batch {batchId} not found")
-    return {"lines": state.log_lines[-200:]}
+    return {"lines": state.log_lines[-2000:]}
 
 
 @router.get("/api/lab/debug-log")

ui/ai-lab.js

@@ -1397,9 +1397,8 @@ const AiLab = (() => {
                 const data = await res.json();
                 const logEl = document.getElementById('lab-log-content');
                 if (logEl && data.lines) {
-                    // Show last N lines
-                    const tail = data.lines.slice(-30);
-                    logEl.textContent = tail.join('\n');
+                    // Show all lines from the backend (up to 200)
+                    logEl.textContent = data.lines.join('\n');
                     logEl.scrollTop = logEl.scrollHeight;
                 }
             } catch { /* ignore */ }