pub fn fp4_quant_e8m0_inplace(x: &mut [f32], block: usize)
fp4_act_quant(x, block, inplace=True) with e8m0 scales: scale = pow2 ceiling of amax / 6 (amax floored at 6 * 2^-126), values to e2m1, dequantised back to bf16.
fp4_act_quant(x, block, inplace=True)
amax / 6
6 * 2^-126