Make ckpool client resilient to warm-up EOFs

The stratifier can accept a connection and close it without
writing a response during early startup — the refresh loop
then logs 'ckpool poolstats failed: read len: EOF' on every
tick until the stratifier is ready, which is noisy and looks
like a real fault.

- client.Send retries once after 200ms if the server closed
  the connection before any bytes were read (io.EOF anywhere
  in the wrapped chain).
- Aggregator tracks a ckFailStreak counter: first two
  consecutive failures log at DEBUG, third and beyond escalate
  to WARN. Successful refreshes reset the streak.

api/internal/ckpool/client.go

@@ -51,7 +51,24 @@ var ErrProtocol = errors.New("ckpool: protocol error")
 // Send delivers a single command to the named ckpool socket and returns the
 // raw response bytes. `sockName` is one of SocketListener, SocketStratifier,
 // or SocketConnector.
+//
+// If the server closes the connection before writing a response (common
+// during ckpool warm-up — the stratifier accepts the socket but its stats
+// aren't populated yet), Send retries once after a short backoff.
 func (c *Client) Send(ctx context.Context, sockName, cmd string) ([]byte, error) {
+	buf, err := c.sendOnce(ctx, sockName, cmd)
+	if err != nil && errors.Is(err, io.EOF) {
+		select {
+		case <-ctx.Done():
+			return nil, ctx.Err()
+		case <-time.After(200 * time.Millisecond):
+		}
+		return c.sendOnce(ctx, sockName, cmd)
+	}
+	return buf, err
+}
+
+func (c *Client) sendOnce(ctx context.Context, sockName, cmd string) ([]byte, error) {
 	if cmd == "" {
 		return nil, fmt.Errorf("ckpool: empty command")
 	}