Support alternative UUID formats.

[jcrist]

This adds a new uuid_format to the json and msgpack encoders, configuring how msgspec will handle uuids. This may be one of:

• canonical: UUIDs are encoded as canonical strings (same as str(uuid)). This is the default.
• hex: UUIDs are encoded as strings without hyphens (same as uuid.hex).
• bytes: UUIDs are encoded as binary values representing big-endian 128-bit integers (same as uuid.bytes).

All protocols support decoding from any of these formats by default. Note that only msgpack supports encoding/decoding the 'bytes' value, as JSON lacks a native binary type.

In [1]: import msgspec

In [2]: import uuid

In [3]: u = uuid.uuid4()

In [4]: u
Out[4]: UUID('827ba710-601d-421c-bdaf-632d2db2620c')

In [5]: msgspec.json.encode(u)  # defaults to canonical
Out[5]: b'"827ba710-601d-421c-bdaf-632d2db2620c"'

In [6]: hex_enc = msgspec.json.Encoder(uuid_format="hex")  # use the hex format

In [7]: hex_enc.encode(u)
Out[7]: b'"827ba710601d421cbdaf632d2db2620c"'

In [8]: msgspec.json.decode(_, type=uuid.UUID)  # decode supports all formats
Out[8]: UUID('827ba710-601d-421c-bdaf-632d2db2620c')

In [9]: bytes_enc = msgspec.msgpack.Encoder(uuid_format="bytes")  # use the bytes format

In [10]: bytes_enc.encode(u)
Out[10]: b'\xc4\x10\x82{\xa7\x10`\x1dB\x1c\xbd\xafc--\xb2b\x0c'

In [11]: u.bytes  # this ^^ is the msgpack encoded version of u.bytes
Out[11]: b'\x82{\xa7\x10`\x1dB\x1c\xbd\xafc--\xb2b\x0c'

In [12]: msgspec.msgpack.decode(bytes_enc.encode(u), type=uuid.UUID)  # roundtrips fine
Out[12]: UUID('827ba710-601d-421c-bdaf-632d2db2620c')

Fixes #493.