Rectify language fields

[porcuquine]

This PR attempts to rectify confusion and inaccuracy wreaking havoc in our naming of fields. This is really important for multiple reasons. The worst-case situation is one in which we accidentally instantiate a Lurk language on the wrong field. This would lead to incompatible data and great confusion. The less bad cases involve conceptual incoherence, false labeling, and general mayhem in the vicinity of our efforts to produce an evidently correct code base.

Please see individual commits for details. Problems fixed here:

• In some places, the wrong scalar field was being used. I believe this problem was introduced in Refactor LurkField with lurk_ff changes #239, then either independently 'discovered' or else cargo-culted by reference in new benchmarks.
• In some places, type aliases which do not obviously match the intended LanguageField were used. This would be fine if the actual correspondences between the pasta_curves named fields (Fq and Fp) were well-known, but apparently it's not. In any case, using the explicative aliases eliminates the need for it to (usually) be. The same goes for use of nova::S1 and nova::S2.
• The finer points of this (while all discoverable by reading code whose base case are the named aliases in pasta_curves) were not explicitly documented outside of code.

If you contribute to lurk-rs, please take note.

[huitseeker]

You can chuck the following in a clippy.toml at the root of the repo to have the yelling done by tooling instead of you:

disallowed-methods = [
    # we use strict naming conventions for fields of our workhorse curve
    { path = "pasta_curves::Fp", reason = "use pasta_curves::pallas::Scalar instead" },
    ...
]

[porcuquine]

Thanks. However, please note that the code you provided here reproduces the error I sought to correct here. In fact, pasta_curves::Fq is the one that corresponds to pallas::Scalar — as noted in the added documentation.

I realize this isn't what you were intending to communicate, and I mention it only because:

• It's still wrong…
• It's a perfect and timely example of how allowing these wrong expressions into our source of truth creates ongoing confusion.

As always, I recognize that I did approve the PR(s) introducing these errors.