This seems absurd as the speed of photon is constant. And according to NP-Tetrad 3 base-null vectors of photons are considered in the metric space. Therefore, if in the region of magnetorotational instability, a photon curves its path either in unison or not in spin direction of the black hole, experiences change in velocity called acceleration? This is also a question of comoving photons with ions.

There's no meaningful difference between ``acceleration'' and ``deceleration''. In a curved spacetime the speed of a photon isn't constant, incidentally, precisely due to the property that spacetime is curved.