Add test and fix for non-canonical point at infinity

[kevaundray]

Description

closes: #XXXX

---

Before we can merge this PR, please make sure that all the following items have been
checked off. If any of the checklist items are not applicable, please leave them but
write a little note why.

• Targeted PR against correct branch (master)
• Linked to Github issue with discussion and accepted design OR have an explanation in the PR that describes this work.
• Wrote unit tests
• Updated relevant documentation in the code
• Added a relevant changelog entry to the Pending section in CHANGELOG.md
• Re-reviewed Files changed in the Github PR explorer

[Pratyush]

Should we make this a byte-based check? i.e. directly on the bytes that we've read?

[mmagician]

Good catch, I've PRed to @kevaundray's fork, if not merged there soon I'll open a PR directly here.

[ArnaudBrousseau]

I think some extra validation here is needed to catch another, more subtle class of non-canonical infinity point encodings: some flag combinations are nonsensical. Here's a failing test to show you what I mean:

#[test]
fn bad_flag_combination() {
    // See https://github.com/zkcrypto/pairing/tree/fa8103764a07bd273927447d434de18aace252d3/src/bls12_381#serialization
    // - Bit 1 is compressed/uncompressed
    // - Bit 2 is infinity
    // - Bit 3 is lexicographical order for compressed point deserialization
    // Hence `0b1110` ("e" in hex) or `0b0110` ("6" in hex") are both nonsensical.
    let non_canonical_hex = "e00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000";
    let non_canonical_bytes = hex::decode(non_canonical_hex).unwrap();
    assert_eq!(non_canonical_bytes.len(), 48);

    let maybe_affine_point: Result<G1Affine, ark_serialize::SerializationError> =
        CanonicalDeserialize::deserialize_compressed(&non_canonical_bytes[..]);

    assert!(maybe_affine_point.is_err())
}