montgomery representation

[TalDerei]

https://github.com/ingonyama-zk/icicle/blob/00231723b9fe0e2d7cbfc3202109cd6399b4c523/icicle/curves/bls12_381.cuh#LL27C41-L27C44

This is usually represented as the montgomery form of 1 if you want to convert to between affine/jacobian/projective forms: https://github.com/ingonyama-zk/icicle/blob/main/icicle/primitives/projective.cuh#L23

[DmytroTym]

This has to do with the answer to the previous issue you opened, namely that we currently don't use Montgomery multiplication. If we succeed in having competitive non-Montgomery multiplier, the config file can stay the way it is. If we don't and we have to move to Montgomery, then you're right, one needs to be one in Montgomery form, and not just plain 1.

[DmytroTym]

I feel like this can be closed due to #289
We are on par with the state-of-the art Montgomery multiplier so it feels like there's no urgency to move to it. Can be reopened if people improve Montgomery multiplication or we just decide to support Montgomery as well.