fix: execute-stream POST → GET für EventSource

Backend:
- prompts.py: @router.post → @router.get für /execute-stream
- EventSource unterstützt nur GET-Requests
- modules/timeframes nutzen Defaults (SSE kann keine komplexen Params)

Fixes:
- "Connection to server lost" bei Analyse-Ausführung

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

backend/routers/prompts.py

@@ -1445,12 +1445,10 @@ from prompt_executor import execute_prompt_with_data
 from models import UnifiedPromptCreate, UnifiedPromptUpdate
 
 
-@router.post("/execute-stream")
+@router.get("/execute-stream")
 async def execute_unified_prompt_stream(
     prompt_slug: str = Query(..., description="Slug of prompt to execute"),
     token: Optional[str] = Query(None, description="Auth token (temporary solution for SSE)"),
-    modules: Optional[dict] = None,
-    timeframes: Optional[dict] = None,
     debug: bool = Query(False, description="Include debug information (node_states, etc.)"),
     save: bool = Query(False, description="Save result to ai_insights")
 ):
@@ -1477,24 +1475,22 @@ async def execute_unified_prompt_stream(
             raise HTTPException(401, "Invalid or expired token")
         profile_id = row['profile_id']
 
-    # Use default modules/timeframes if not provided
-    if not modules:
-        modules = {
-            'körper': True,
-            'ernährung': True,
-            'training': True,
-            'schlaf': True,
-            'vitalwerte': True
-        }
+    # Use default modules/timeframes (SSE doesn't support complex params)
+    modules = {
+        'körper': True,
+        'ernährung': True,
+        'training': True,
+        'schlaf': True,
+        'vitalwerte': True
+    }
 
-    if not timeframes:
-        timeframes = {
-            'körper': 30,
-            'ernährung': 30,
-            'training': 14,
-            'schlaf': 14,
-            'vitalwerte': 7
-        }
+    timeframes = {
+        'körper': 30,
+        'ernährung': 30,
+        'training': 14,
+        'schlaf': 14,
+        'vitalwerte': 7
+    }
 
     # Wrapper function for OpenRouter calls
     async def workflow_llm_call(prompt: str, model: str = None) -> str: