Add fake redis server to properly test connection handshake

[kristjanvalur]

Pull Request check-list

Please make sure to review and check all of these items:

• Do tests and lints pass with this change?
• Do the CI tests pass with this change (enable it first in your forked repo and wait for the github action build to finish)?
• Is the new or changed code fully tested?
• Is a documentation update included (if this change modifies existing APIs, or introduces new ones)?
• Is there an example added to the examples folder (if applicable)?
• Was the change added to CHANGES file?

Description of change

This PR provides a transport agnostic RESP parser and encoder for the unit tests.
It also provides a simple, transport agnostic, RESP server. This allows us to share code and logic
between sync and async code.
A parametrized test is added to test the connection handshake for different client settings against different fake REDIS servers,
both sync and async.

• the RESP parser / encoder is simple, assumes utf8 encoding, and always does decoding (from bytes to string.)
• Decoding uses the 'surrogateescape' convention, so that if binary data was intended (a blob) it can always be recovered by re-encoding.
• Decoding supports most RESP 3 data types
• Encoding assumes utf8, and chooses RESP3 types if protocol 3 is enabled, when appropriate.
• RESP server receives resp command arrays and has handlers to respond. Can be expanded to perform more extensive Redis emulation
• RESP server can be plugged into a sync or async request handler to simplify testing.

[codecov-commenter]

Codecov Report

Attention: 35 lines in your changes are missing coverage. Please review.

Comparison is base (0113034) 91.32% compared to head (5a79252) 91.56%.

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

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2947      +/-   ##
==========================================
+ Coverage   91.32%   91.56%   +0.24%     
==========================================
  Files         126      128       +2     
  Lines       32719    33258     +539     
==========================================
+ Hits        29881    30454     +573     
+ Misses       2838     2804      -34     

Files	Coverage Δ
tests/test_resp.py	100.00% <100.00%> (ø)
tests/test_asyncio/test_connect.py	96.03% <91.66%> (-2.07%)	⬇️
tests/test_connect.py	95.45% <91.66%> (-1.54%)	⬇️
tests/resp.py	92.46% <92.46%> (ø)

... and 6 files with indirect coverage changes

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

[Fogapod]

why no type hint for errorhander here?

[Fogapod]

maybe do self.consumed.clear() here to make it more readable? same for all other places

[kristjanvalur]

fair point

[Fogapod]

maybe use CRNL variable here and in all other lines below since its already defined?

[kristjanvalur]

no, CRNL is a bytes object, just a shorthand for parsing, not emitting where we use str.

[Fogapod]

sorry, i commented on wrong line. there are still a few insurances of b"\r\n" in code in addition to strings

[kristjanvalur]

Yes, I like it like that since it makes the code more consistent and explicit where we are writing RESP responses.