How robust are in situ observations for validating satellite‐derived albedo over the dark zone of the Greenland Ice Sheet?