fix: tool registry emits Anthropic schema, question lifecycle cleanup

- Tool.to_schema() emits input_schema (Anthropic Messages API) instead
  of OpenAI-style parameters
- QuestionManager: remove questions from _pending on answer and timeout,
  reject double answers, add list_pending()
- core/skills/__init__.py so skills are importable as a package
- revert optional-import fallbacks in logger/message_bus: nats-py and
  loguru are hard requirements, a broken install should fail loudly
- tests for wait_for_answer happy path and timeout cleanup

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012AJLYcbTHCe45fyFnMfEin

core/tool_registry.py

@@ -33,7 +33,7 @@ class Tool:
     metadata: Dict[str, Any] = field(default_factory=dict)
 
     def to_schema(self) -> Dict[str, Any]:
-        """Generate OpenAI/Claude compatible JSON Schema for tool calling."""
+        """Generate an Anthropic Messages API tool definition (name / description / input_schema)."""
         properties = {}
         required = []
 
@@ -56,7 +56,7 @@ class Tool:
         return {
             "name": self.name,
             "description": self.description,
-            "parameters": {
+            "input_schema": {
                 "type": "object",
                 "properties": properties,
                 "required": required,