Error messages are not handled correctly when using batchcall if the server does not support it.

[EthanShang8989]

// BatchCallContext sends all given requests as a single batch and waits for the server
// to return a response for all of them. The wait duration is bounded by the
// context's deadline.
func (c *Client) BatchCallContext(ctx context.Context, b []BatchElem) error {
	var (
		msgs = make([]*jsonrpcMessage, len(b))
		byID = make(map[string]int, len(b))
	)
	for i, elem := range b {
		msg, err := c.newMessage(elem.Method, elem.Args...)
		if err != nil {
			return err
		}
		msgs[i] = msg
		byID[string(msg.ID)] = i
	}

	respBody, err := c.doRequest(ctx, msgs)
	if err != nil {
		return err
	}
	defer respBody.Close()

**//Need to add error handling here**

	var respmsgs []jsonrpcMessage
	if err := json.NewDecoder(respBody).Decode(&respmsgs); err != nil {
		return err
	}

	for idx, resp := range respmsgs {
		elem := &b[idx]
		if resp.Error != nil {
			elem.Error = resp.Error
			continue
		}
		if len(resp.Result) == 0 {
			elem.Error = ErrNoResult
			continue
		}
		elem.Error = json.Unmarshal(resp.Result, elem.Result)
	}
	return nil
}