Is magnetic resonance spectroscopy capable of detecting metabolic abnormalities in neurofibromatosis type 1 that are not revealed in brain parenchyma of normal appearance?