Implemented the binary wire format

[Hydrocharged]

Fixes:

• Extended protocol SELECT fails with 'permission denied for table pg_type' #2317
• Extended protocol sends empty binary payload for NUMERIC/DECIMAL columns, causing client panic #2347
• Extended protocol sends incorrect binary format for ENUM columns, causing client-side decode panic #2348
• Extended protocol sends JSONB as raw JSON text instead of binary JSONB format (missing 0x01 version prefix) #2360
• COUNT(*) returns wrong value via binary protocol (extended query) #2362

In a nutshell, we did not respect a client's wishes when they requested that the results for some columns are returned in their binary format (obtained from their send function). Instead, we had hardcoded that some types, in certain circumstances, would only return their binary format, while others would strictly return their text format. This PR implements the binary wire format for every type that is currently supported. This also fixes numerous bugs that were hidden due to our incorrect behavior.

In addition to the above fixes, this PR introduces a new test harness for analyzing our compatibility with the wire messages returned directly to Postgres clients. Thousands of lines of new tests have been added that ensure we're returning the exact correct results for a given set of messages (not just queries but the actual client messages), and the test harness allows for quick iteration when comparing what a Postgres server sends versus a Doltgres server, so we can ensure that we're returning the same bytes, not just the same overall structure.

[github-actions]

	Main	PR
covering_index_scan_postgres	554.61/s	586.65/s	+5.7%
index_join_postgres	120.65/s	124.32/s	+3.0%
index_join_scan_postgres	176.22/s	182.69/s	+3.6%
index_scan_postgres	11.15/s	11.06/s	-0.9%
oltp_point_select	2356.77/s	2367.29/s	+0.4%
oltp_read_only	1679.19/s	1693.84/s	+0.8%
select_random_points	110.44/s	112.02/s	+1.4%
select_random_ranges	434.23/s	443.97/s	+2.2%
table_scan_postgres	10.87/s	10.87/s	0.0%
types_table_scan_postgres	4.93/s	5.00/s	+1.4%

[github-actions]

	Main	PR
Total	42090	42090
Successful	17648	17651
Failures	24442	24439
Partial Successes1	5575	5575

	Main	PR
Successful	41.9292%	41.9363%
Failures	58.0708%	58.0637%

${\color{lightgreen}Progressions (3)}$

psql

QUERY: SELECT 1 AS one ; SELECT warn('1.5') ; SELECT 2 AS two ;

QUERY: SELECT 3 AS three ; SELECT warn('3.5') ; SELECT 4 AS four 

QUERY: SELECT 1 AS one ; SELECT warn('1.5') ; SELECT 2 AS two ;

Footnotes

1. These are tests that we're marking as Successful, however they do not match the expected output in some way. This is due to small differences, such as different wording on the error messages, or the column names being incorrect while the data itself is correct. ↩

[fulghum]

Looks great. Very cool that we can easily run the new test harness against a Postgres server and confirm that our expected wire messages are correct.

[fulghum]

A customer pinged us about needing this change, so I'm going to go ahead and merge it in and get a release out for them.