fix(api): skip preflight for unknown model limits

rust/crates/api/src/providers/mod.rs

@@ -640,14 +640,7 @@ pub fn model_token_limit(model: &str) -> Option<ModelTokenLimit> {
             max_output_tokens: 16_384,
             context_window_tokens: 256_000,
         }),
-        // Hotfix: Unknown models get conservative defaults to avoid crashes.
+        _ => None,
-        // Uses the minimum of known supported models: max_output_tokens: 16_384,
-        // context_window_tokens: 131_072. This may under-utilize the model's
-        // actual capabilities but hopefully ensures safer operation.
-        _ => Some(ModelTokenLimit {
-            max_output_tokens: 16_384,
-            context_window_tokens: 131_072,
-        }),
     }
 }