feat: wire up federation to use peer swarms for generation

- Modify /v1/chat/completions endpoint to check for federation
- If federation enabled with peers, use generate_with_federation()
- Otherwise fall back to local generation
- Add --peer example to help text

Now when federation is enabled and peers are discovered/manually added,
generation requests will be distributed across local and peer swarms,
with consensus voting to select the best response.

main.py

@@ -123,6 +123,7 @@ Examples:
   python main.py --test             # Test with sample prompt
   python main.py --mcp              # Enable MCP server
   python main.py --federation       # Enable federation with other instances
+  python main.py --federation --peer 192.168.1.10:17615  # Manual peer
         """
     )
     

src/api/routes.py

@@ -297,6 +297,46 @@ async def chat_completions(request: ChatCompletionRequest):
     else:
         # Regular response with consensus
         try:
+            # Use federation if enabled and peers are available
+            if federated_swarm is not None:
+                peers = federated_swarm.discovery.get_peers()
+                if peers:
+                    print(f"🌐 Using federation with {len(peers)} peer(s)...")
+                    result = await federated_swarm.generate_with_federation(
+                        prompt=prompt,
+                        max_tokens=request.max_tokens or 1024,
+                        temperature=request.temperature or 0.7,
+                        min_peers=0  # Allow local fallback if no peers respond
+                    )
+                    response_text = result.final_response
+                    tokens_generated = len(response_text.split())  # Rough estimate
+                    
+                    # Estimate prompt tokens (rough approximation)
+                    prompt_tokens = len(prompt) // 4
+                    
+                    return ChatCompletionResponse(
+                        id=completion_id,
+                        created=created,
+                        model=request.model,
+                        choices=[
+                            ChatCompletionChoice(
+                                index=0,
+                                message=ChatMessage(
+                                    role="assistant",
+                                    content=response_text,
+                                    tool_calls=[]
+                                ),
+                                finish_reason="stop"
+                            )
+                        ],
+                        usage=UsageInfo(
+                            prompt_tokens=prompt_tokens,
+                            completion_tokens=tokens_generated,
+                            total_tokens=prompt_tokens + tokens_generated
+                        )
+                    )
+            
+            # Fallback to local generation
             result = await swarm_manager.generate(
                 prompt=prompt,
                 max_tokens=request.max_tokens or 1024,