fix(models): query v1 models for llama-server endpoints (#3380)

* fix(models): query v1 models for llama-server endpoints

* test(models): accept owner kwargs in llama-server regression

src/endpoint_resolver.py

@@ -184,7 +184,7 @@ def build_chat_url(base: str) -> str:
 
 def build_models_url(base: str) -> Optional[str]:
     """Return the provider-specific model-list endpoint URL for a base."""
-    base = resolve_url(base)
+    base = normalize_base(resolve_url(base))
     provider = _detect_provider(base)
     if provider == "anthropic":
         return _anthropic_api_root(base) + "/v1/models"

src/llm_core.py

@@ -1042,7 +1042,9 @@ def list_model_ids(
         if provider == "ollama":
             models_url = _ollama_api_root(base_chat_url) + "/tags"
         else:
-            models_url = base_chat_url.replace("/chat/completions", "/models")
+            from src.endpoint_resolver import build_models_url
+
+            models_url = build_models_url(base_chat_url)
         r = httpx.get(models_url, headers=h, timeout=timeout)
         r.raise_for_status()
         data = r.json()

src/model_context.py

@@ -297,7 +297,9 @@ def _query_context_length(endpoint_url: str, model: str) -> int:
             logger.info(f"Using known context window for {model}: {known}")
         return known or DEFAULT_CONTEXT
 
-    models_url = endpoint_url.replace("/chat/completions", "/models")
+    from src.endpoint_resolver import build_models_url
+
+    models_url = build_models_url(endpoint_url)
     try:
         r = httpx.get(models_url, timeout=REQUEST_TIMEOUT)
         if r.is_success: