fix: Add wrapper function for workflow LLM calls to prevent max_tokens signature mismatch

Problem: workflow_executor calls openrouter_call_func(prompt, model) but
call_openrouter expects (prompt, max_tokens=4096). This caused the model string
'anthropic/claude-sonnet-4' to be passed as max_tokens, resulting in OpenRouter
requesting 64000 tokens and failing with 402 credit errors.

Solution: Added workflow_llm_call() wrapper in workflows.py that matches the
expected (prompt, model) -> str signature and calls call_openrouter correctly.

Fixes: All workflows failing with 402 'insufficient credits' errors

backend/routers/workflows.py

@@ -86,11 +86,19 @@ async def execute_workflow_endpoint(
             "type": "workflow"
         }
 
+    # Wrapper function to match workflow_executor's expected signature: (prompt, model) -> str
+    # workflow_executor calls: openrouter_call_func(prompt, "anthropic/claude-sonnet-4")
+    # but call_openrouter expects: call_openrouter(prompt, max_tokens=4096)
+    async def workflow_llm_call(prompt: str, model: str) -> str:
+        # Ignore model parameter (already set in OPENROUTER_MODEL env var)
+        # Use default max_tokens=4096 from call_openrouter
+        return await call_openrouter(prompt)
+
     try:
         result = await execute_workflow_prompt(
             prompt=workflow_prompt,
             variables=variables,
-            openrouter_call_func=call_openrouter,
+            openrouter_call_func=workflow_llm_call,  # Use wrapper with correct signature
             enable_debug=request.enable_debug
         )
         return result