This was more useful at the time when we were determining, e.g., optimal lookup
table sizes and could regenerate them more easily, but it came at a massive
complexity cost. It also meant that we were unable to implement backend
autoselection. This commit removes the `build.rs` entirely. In the future, a
different `build.rs` could be added that auto-selects a backend, but it seems
like the current default-u64 setup has been working fine.
Since computation of the 0 term in reduction requires a multiplication with a
4-cycle latency, this ensures that the rest of the computation can start before
the 0 term is finished
This splits the `FieldElement51x4` type into two types:
- `F51x4Reduced` (with reduced limbs)
- `F51x4Unreduced` (with unreduced limbs)
The reduction is implemented as a `From` impl to convert one type to the other.
The output of a multiplication is now a `F51x4Unreduced`. The reason is that
the inputs to IFMA operations must be at most 52 bits, so it's not possible to
perform an addition of (51+epsilon)-bit values and still be small enough to be
used as an input to multiplication. So, it doesn't make sense to perform a
reduction at the end of a multiplication, because the reduced values will be
fed into an addition or subtraction, which then needs to be re-reduced.
