Probing the "Psyche'' of Large Reasoning Models: Understanding Through a Human Lens