Change from_bytes methods to take fixed-size array argument

[mina86]

Change from_bytes methods of all the types to take &[u8; N] argument
(with N appropriate for given type) rather than &[u8]. This does
a few things:

• it makes SecretKey::from_bytes infallible which will simplify call
  sites (if the call site already has properly-sized array),
• harmonises the convention with ed25519::Signature::from_bytes method
  and
• encodes the size of the bytes representation in type system which
  means that users can assert it at compile time.

To still allow creating the objects from a slice whose size is checked
at run time, introduce TryFrom<&[u8]> implementations for affected
public types.

This is an API breaking change. The simplest way to update existing
code is to replace Foo::from_bytes with Foo::try_from. This should
cover majority of uses.

[tarcieri]

In general I'd suggest holding off on this until #225 is addressed, since that proposes to change the public APIs for keys

[tarcieri]

This was also an interesting read on the actual costs of length checks: https://blog.readyset.io/bounds-checks/

tl;dr: not much, especially in cases like this

[mina86]

The main thing is that creation of SecretKey is now infallible. Another is that the expected length is now expressed in type system. Performance isn’t really my main concern.

[tarcieri]

I'd probably suggest making the TryFrom impls take &[u8] and using inherent methods (i.e. from_bytes) for the fixed-sized inputs. That's the approach we use with @RustCrypto.

[mina86]

I can change from_bytes if you prefer though that is an API breaking change.

[mina86]

So I’m looking at ed25519::Signature::from_bytes from a preview ed25519 crate and it has from_bytes(&[u8]) so should this follow that instead? Or should ed25519 crate be changed as well?

[tarcieri]

I'd be open to changing that method

[mina86]

Updated as per suggestion. Happy to wait till #225 gets resolved.

[tarcieri]

#225 is resolved

[tarcieri]

@mina86 you need to change the base to release/2.0