ark_std::error vs core::error

[burdges]

ark_std::error replaces core::error when not using std, but it winds up woefully inferior to the core::error behind #[feature(error_in_core)]. In particular ark_std::error::Error should've the downcast* methods on core::error::Error.

We should probably copy core::error to ark_std::error. Ain't much relevant in the real std::error anymore. If possible, we should detect #[feature(error_in_core)] or even nightly, so as to not replace core::error then.

Interestingly, there were never std::io references in std::error per se. It appears backtrace messed up moving std::error to core::error, but I've no idea why backtrace cannot work with just alloc, or be placed into an extension trait, or whatever.

[burdges]

See w3f/ring-vrf@b87e3a1#diff-b956979688cec9dec3a8d6eeb4ccd2a129e0d48803110ee9def443aec17d72d7R105

[Pratyush]

100% agree that we should switch to core::error, but as things stand this is not actionable, right? We need error_in_core to be stabilized first. I would prefer to avoid rust version detection, as that would involve a build.rs which worsens compile time.

[burdges]

I simply meant copy "more of" std::error from std into ark-std, specifically the <dyn Error>::downcast part.

I'll make @achimcc submit a PR for it if our auditors think error passing matters in ark-scale link above. We've correct error passing in the substrate host already, which maybe enough.