Remove PSQLJSONEncoder

[fabianfett]

Motivation

While rewriting lot's of PostgresNIO internals in the last year, I created new types that very closely match existing types. This was a mistake. I should have tried everything to reuse existing types, and change them as needed.

Changes

⚠️ This PR extends API... minor version bump required.

• Remove PSQLJSONEncoder
• Extend PostgresJSONEncoder with an encode method based on ByteBuffer
• Provide a default implementation for the new encode method based on ByteBuffer to not break API
• Replace all uses of PSQLJSONEncoder with PostgresJSONEncoder

Result

Fewer types, easier API for adopters.

[codecov-commenter]

Codecov Report

Merging #215 (47281a6) into main (eaef208) will increase coverage by 4.17%.
The diff coverage is 37.50%.

@@            Coverage Diff             @@
##             main     #215      +/-   ##
==========================================
+ Coverage   40.69%   44.87%   +4.17%     
==========================================
  Files         117      118       +1     
  Lines        7709     7713       +4     
==========================================
+ Hits         3137     3461     +324     
+ Misses       4572     4252     -320     

Flag	Coverage Δ
unittests	44.87% <37.50%> (+4.17%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...resNIO/Connection/PostgresConnection+Connect.swift	0.00% <0.00%> (ø)
Sources/PostgresNIO/New/PSQLCodable.swift	100.00% <ø> (ø)
Sources/PostgresNIO/Postgres+PSQLCompat.swift	0.00% <ø> (ø)
...es/PostgresNIO/Utilities/PostgresJSONEncoder.swift	0.00% <0.00%> (ø)
Sources/PostgresNIO/New/Messages/Bind.swift	100.00% <100.00%> (ø)
Sources/PostgresNIO/New/PSQLConnection.swift	31.68% <100.00%> (+1.98%)	⬆️
...s/PostgresNIO/New/PSQLFrontendMessageEncoder.swift	100.00% <100.00%> (ø)
...urces/PostgresNIO/New/Data/Array+PSQLCodable.swift	94.68% <0.00%> (+1.06%)	⬆️
...rces/PostgresNIO/New/Extensions/Logging+PSQL.swift	19.60% <0.00%> (+1.42%)	⬆️
... and 25 more

[gwynne]

Same as #212 and #213, why are you wasting my time reading my approval instead of merging lovely code cleanups faster than I can thumbs-up 'em? 🙃