Fix to_bytes() for ProjectivePoint::identity()

[daxpedda]

This could maybe alternatively be fixed "upstream" in EncodedPoint::len()?
Fixes #442.

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

Attention: Patch coverage is 71.42857% with 2 lines in your changes missing coverage. Please review.

Project coverage is 64.51%. Comparing base (2d91c16) to head (0ba5440).
Report is 640 commits behind head on master.

Files with missing lines	Patch %	Lines
p256/src/arithmetic/projective.rs	71.42%	2 Missing ⚠️

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #443      +/-   ##
==========================================
+ Coverage   64.45%   64.51%   +0.05%     
==========================================
  Files          28       28              
  Lines        3584     3590       +6     
==========================================
+ Hits         2310     2316       +6     
  Misses       1274     1274              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[tarcieri]

This is an unfortunate mismatch between the GroupEncoding API and SEC1 encoding.

This could maybe alternatively be fixed "upstream" in EncodedPoint::len()?

EncodedPoint::len is doing the right thing: the SEC1 encoding for the identity point is a 1-byte 0 tag.

Technically this is an invalid SEC1 encoding, but there's nothing else that can be done with a fixed-sized GroupEncoding::Repr, and it's certainly preferable to panicking.

[daxpedda]

Technically this is an invalid SEC1 encoding, but there's nothing else that can be done with a fixed-sized GroupEncoding::Repr, and it's certainly preferable to panicking.

We could change the interface to return an Option?

[tarcieri]

GroupEncoding is defined in the group crate, so the change would have to be made there.

Really it's an eccentricity of the SEC1 encoding, and I doubt they would want to make the API fallible just to support that.

[daxpedda]

GroupEncoding is defined in the group crate, so the change would have to be made there.

Ah, I wasn't aware this isn't owned by RustCrypto.

[tarcieri]

It'd also be good to make sure that GroupEncoding::from_bytes can decode this encoding, and also that the problem is also solved for the GroupEncoding impl on AffinePoint.

[tarcieri]

This is fine as a start. I'll take care of the other comments I raised.