fix: backend improvements (customer health, tests, middleware, corpus)

backend/app/services/corpus_trainer.py

@@ -1,5 +1,5 @@
 from typing import Dict, Any, Optional, List
-from sqlalchemy import select, func, and_
+from sqlalchemy import select, func, and_, String
 from sqlalchemy.ext.asyncio import AsyncSession
 from datetime import datetime, timedelta
 import logging
@@ -58,7 +58,7 @@ class CorpusTrainer:
             select(
                 CorpusEntry.source_text,
                 CorpusEntry.task_type,
-                func.min(CorpusEntry.id).label("keep_id"),
+                func.min(CorpusEntry.id.cast(String)).label("keep_id"),
             )
             .group_by(CorpusEntry.source_text, CorpusEntry.task_type)
             .having(func.count(CorpusEntry.id) > 1)
@@ -70,7 +70,7 @@ class CorpusTrainer:
                 and_(
                     CorpusEntry.source_text == subquery.c.source_text,
                     CorpusEntry.task_type == subquery.c.task_type,
-                    CorpusEntry.id != subquery.c.keep_id,
+                    CorpusEntry.id.cast(String) != subquery.c.keep_id,
                 )
             )
         )
@@ -170,7 +170,7 @@ class CorpusTrainer:
             from app.config import settings
             import httpx
 
-            if settings.OPENAI_API_KEY:
+            if getattr(settings, 'OPENAI_API_KEY', None):
                 async with httpx.AsyncClient() as client:
                     resp = await client.post(
                         "https://api.openai.com/v1/embeddings",