Support for using TEXT as a foreign key target

[zachmu]

This PR relies on dolthub/go-mysql-server#2882

[github-actions]

	Main	PR
covering_index_scan_postgres	378.06/s	380.19/s	+0.5%
index_join_postgres	146.20/s	148.01/s	+1.2%
index_join_scan_postgres	179.57/s	180.95/s	+0.7%
index_scan_postgres	12.13/s	12.16/s	+0.2%
oltp_point_select	2730.27/s	2763.19/s	+1.2%
oltp_read_only	1819.80/s	1810.59/s	-0.6%
select_random_points	104.32/s	104.82/s	+0.4%
select_random_ranges	131.28/s	130.78/s	-0.4%
table_scan_postgres	10.01/s	10.10/s	+0.8%
types_table_scan_postgres	5.27/s	5.31/s	+0.7%

[github-actions]

	Main	PR
Total	42090	42090
Successful	15693	15694
Failures	26397	26396
Partial Successes1	5201	5201

	Main	PR
Successful	37.2844%	37.2868%
Failures	62.7156%	62.7132%

${\color{lightgreen}Progressions (1)}$

triggers

QUERY: create table convslot_test_child (col1 text primary key,
	foreign key (col1) references convslot_test_parent(col1) on delete cascade on update cascade
);

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. ↩

[Hydrocharged]

These can be DoltgresTypes rather than sql.Type

[Hydrocharged]

Adding on to my other comment, we aren't too strict on ensuring that these are returned correctly, considering they don't map to all of our types so it's not really possible to do so. Instead, it's better to check for these specific types. You could even use type1.ID.TypeName() and switch on the string name. But like oid returns sqltypes.VarChar cause there's no oid type in MySQL so we're just returning something.

I think switch type1.ID with case pgtypes.Varchar.ID should work, but if not you can grab the case values using Id.CaseString() and match on the string directly (so you'd just use case "CaseString_Output").

[zachmu]

Subsumed by #1277