infospace: process book-1-chapter-03 with LLM integration

Auto-generated mappings and analysis via Claude Code CLI adapter.
Entities were already present from a previous session.

Stats: 5m04s wall time, ~51K estimated tokens, ~$0.35 estimated cost.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

examples/infospace-with-history/process_chapters.py

@@ -254,15 +254,26 @@ class ChapterProcessor:
 
         Returns the generated content, or ``None`` on failure.
         """
+        import time as _time
         from markitect.prompts.execution.models import RunConfig
 
         print(f"        Calling LLM ({stage_label})...")
+        t0 = _time.time()
         try:
             response = self.llm_adapter.execute_prompt(prompt, RunConfig())
         except Exception as exc:
-            print(f"        LLM error: {exc}")
+            print(f"        LLM error ({_time.time() - t0:.1f}s): {exc}")
             return None
 
+        elapsed = _time.time() - t0
+        usage = response.usage
+        print(
+            f"        LLM done in {elapsed:.1f}s — "
+            f"prompt {usage.get('prompt_tokens', '?')} tok, "
+            f"completion {usage.get('completion_tokens', '?')} tok, "
+            f"total {usage.get('total_tokens', '?')} tok"
+        )
+
         content = response.content
         if not content or not content.strip():
             print(f"        LLM returned empty content")