fix: Correct OpenAI CLI syntax for chat completions

- Fix openai CLI command to use correct syntax: chat.completions.create
- Use -g flag instead of -p for message input
- Bump version to 0.3.16

The correct command is now:
openai api chat.completions.create -m gpt-4 -g 'message'

pkg/hanzo/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "hanzo"
-version = "0.3.15"
+version = "0.3.16"
 description = "Hanzo AI - Complete AI Infrastructure Platform with CLI, Router, MCP, and Agent Runtime"
 authors = [
     {name = "Hanzo AI", email = "[email protected]"},

pkg/hanzo/src/hanzo/cli.py

@@ -26,7 +26,7 @@ from .utils.output import console
 from .interactive.repl import HanzoREPL
 
 # Version
-__version__ = "0.3.15"
+__version__ = "0.3.16"
 
 
 @click.group(invoke_without_command=True)

pkg/hanzo/src/hanzo/dev.py

@@ -1100,8 +1100,8 @@ Examples:
                 console.print("Then use: hanzo dev --orchestrator codex")
                 return
             
-            # Use openai CLI to chat
-            cmd = ["openai", "api", "chat", "-m", "gpt-4", "-p", message]
+            # Use openai CLI to chat - correct syntax
+            cmd = ["openai", "api", "chat.completions.create", "-m", "gpt-4", "-g", message]
             
             process = subprocess.Popen(
                 cmd,
@@ -2336,7 +2336,7 @@ class MultiClaudeOrchestrator(HanzoDevOrchestrator):
         try:
             import subprocess
             result = subprocess.run(
-                ["openai", "api", "chat", "-m", "gpt-4", "-p", prompt],
+                ["openai", "api", "chat.completions.create", "-m", "gpt-4", "-g", prompt],
                 capture_output=True,
                 text=True,
                 timeout=30