fix(agent): use loguru instead of stdlib logging in runner

The project consistently uses loguru across all 32 modules. The
stdlib logging import was an oversight from the cached_tokens feature.
Also fixes log format from %d/%s placeholders to {} for loguru.

nanobot/agent/runner.py

@@ -3,17 +3,16 @@
 from __future__ import annotations
 
 import asyncio
-import logging
 from dataclasses import dataclass, field
 from typing import Any
 
+from loguru import logger
+
 from nanobot.agent.hook import AgentHook, AgentHookContext
 from nanobot.agent.tools.registry import ToolRegistry
 from nanobot.providers.base import LLMProvider, ToolCallRequest
 from nanobot.utils.helpers import build_assistant_message
 
-logger = logging.getLogger(__name__)
-
 _DEFAULT_MAX_ITERATIONS_MESSAGE = (
     "I reached the maximum number of tool call iterations ({max_iterations}) "
     "without completing the task. You can try breaking the task into smaller steps."
@@ -110,7 +109,7 @@ class AgentRunner:
             context.response = response
             context.usage = iter_usage
             logger.debug(
-                "LLM usage: prompt=%d completion=%d cached=%s",
+                "LLM usage: prompt={} completion={} cached={}",
                 iter_usage["prompt_tokens"],
                 iter_usage["completion_tokens"],
                 iter_usage.get("cached_tokens", 0),