Pace provider traffic and identify genuine foreground bank requests

internal/classification/client.go

@@ -49,7 +49,8 @@ func (c *Client) rateControl() *ratelimit.Controller {
 	if gate := c.rate.Load(); gate != nil {
 		return gate
 	}
-	gate := &ratelimit.Controller{}
+	// Conservative 20-RPM ceiling, independent of model/provider quota claims.
+	gate := &ratelimit.Controller{MinimumInterval: 3 * time.Second, InitialBackoff: 15 * time.Second}
 	if c.rate.CompareAndSwap(nil, gate) {
 		return gate
 	}