Fix double /v1 path when routing to LM Studio

call_openai() (httpx-based) appends /v1/chat/completions to the upstream
URL. But base_url in the models table typically ends in /v1 (matching the
OpenAI SDK convention used by the resolution job). Combining them produced
/v1/v1/chat/completions → 404 from LM Studio.

Strip a trailing /v1 from the stored base_url before passing it to
call_openai() in the agent routing path.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

plugins/festinger/festinger/main.py

@@ -727,7 +727,13 @@ async def _route_agent_chat(
             "content-type": "application/json",
         }
     else:
-        oai_upstream = agent_model.base_url or cfg.get("upstream_openai", "https://api.openai.com")
+        # call_openai() appends /v1/chat/completions to whatever upstream it gets.
+        # DB base_url often ends in /v1 (correct for the OpenAI SDK used by the
+        # resolution job), so strip that suffix to avoid /v1/v1/chat/completions.
+        raw_base = agent_model.base_url or cfg.get("upstream_openai", "https://api.openai.com")
+        oai_upstream = raw_base.rstrip("/")
+        if oai_upstream.endswith("/v1"):
+            oai_upstream = oai_upstream[:-3]
         oai_headers = {
             "authorization": f"Bearer {agent_model.api_key or 'lm-studio'}",
             "content-type": "application/json",