Add agent interface and model implementations

Implements Phase 1: Agent abstraction layer with multiple model support

Features:
- Abstract AgentInterface base class with standard contract
- LoremTesterAgent: Full-featured mock agent with realistic behavior
  - Configurable reasoning effort levels (none to xhigh)
  - Random tool/function call generation
  - Error triggers for testing (rate_limit, context_overflow)
  - Temperature-based response variation
- TatlockAgent: Placeholder for future PydanticAI integration
- ModelRegistry: Centralized model management and discovery

Testing:
- 9 unit tests for lorem-tester agent behavior
- 9 unit tests for registry operations
- Coverage: Agent abstraction fully tested

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

src/agents/tatlock.py

@@ -0,0 +1,78 @@
+"""
+Tatlock agent - Placeholder for future real agent.
+
+This is a minimal placeholder implementation. In the future, this will
+be the production agent using PydanticAI and Ollama for real LLM inference.
+
+For now, it returns a simple placeholder message to show up in the
+model list and allow basic testing.
+"""
+
+import secrets
+from typing import AsyncGenerator, Any
+
+from src.agents.base import AgentInterface, OutputItem
+
+
+def generate_id() -> str:
+    """Generate unique ID for output items."""
+    return secrets.token_hex(16)
+
+
+class TatlockAgent(AgentInterface):
+    """
+    Placeholder for future Tatlock reasoning agent.
+
+    TODO: Integrate PydanticAI and Ollama for real LLM inference
+    TODO: Implement memory modules
+    TODO: Implement expert modules
+    TODO: Add reasoning/thinking capabilities
+    TODO: Add tool/function calling
+    """
+
+    async def generate_response(
+        self,
+        messages: list[dict],
+        reasoning: dict | None = None,
+        tools: list[dict] | None = None,
+        temperature: float = 1.0,
+        max_tokens: int | None = None,
+        stop: list[str] | None = None,
+        **kwargs: Any
+    ) -> AsyncGenerator[OutputItem, None]:
+        """
+        Generate minimal placeholder response.
+
+        In the future, this will call PydanticAI with Ollama backend.
+        """
+
+        # Simple placeholder message
+        yield OutputItem(
+            type="message",
+            id=f"msg_{generate_id()}",
+            role="assistant",
+            content=[{
+                "type": "output_text",
+                "text": "Tatlock agent is not yet implemented. Please use lorem-tester for testing.",
+                "annotations": []
+            }],
+            status="completed"
+        )
+
+    async def supports_tools(self) -> bool:
+        """Tools not yet implemented."""
+        return False
+
+    async def supports_reasoning(self) -> bool:
+        """Reasoning not yet implemented."""
+        return False
+
+    async def get_capabilities(self) -> dict:
+        """Return minimal capabilities."""
+        return {
+            "streaming": True,   # Basic streaming works
+            "reasoning": False,  # Not yet implemented
+            "tools": False,      # Not yet implemented
+            "vision": False,     # Future
+            "audio": False,      # Future
+        }