match mysql fk name generation

[jycor]

This PR removes logic where we would use a consistent hash to create foreign key names, when one isn't explicitly provided.
Instead, we use logic in GMS to generate a name with the template "<tbl_name>_ibfk_<num>", which matches MySQL behavior.

As a result, we now allow multiple foreign keys over the same sets of columns (as long as the names are different).

companion pr:
dolthub/go-mysql-server#2438

fixes:
#7650

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
c48d7a7	ok	5937457

version	total_tests
c48d7a7	5937457

correctness_percentage
100.0

[coffeegoddd]

@coffeegoddd DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
e1ef2eb	ok	5937457

version	total_tests
e1ef2eb	5937457

correctness_percentage
100.0

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
118158c	ok	5937457

version	total_tests
118158c	5937457

correctness_percentage
100.0

[fulghum]

Looks good!

It's a bummer to lose the ability for two branches to both add an FK constraint or secondary index to the same table and merge them back cleanly – that is a cool trick. I think being MySQL compatible ultimately wins out here though, especially since we are giving customers a new statement to easily rename a constraint, and since customers can explicitly specify constraint names to get around the merge conflict issue.

We may eventually want to support both modes, but that's adding complexity for users to deal with, so I'd hold off on that until we know it's definitely a customer problem we need to work on.

Tagging @zachmu on this one, since he may have an opinion on the tradeoff between fully matching MySQL's behavior and keeping the deterministic fk / secondary index naming. Since the customer isn't blocked on this currently, let's give Zach a day to share an opinion, just in case he sees another problem this could cause that we aren't seeing.

[zachmu]

LGTM, see comments on GMS change

[zachmu]

But write tests of merge logic (can skip for now)

If two sides of the merge declare the same fk name with a different definition, that should be a schema conflict.

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
bfd7282	ok	5937457

version	total_tests
bfd7282	5937457

correctness_percentage
100.0

[github-actions]

@coffeegoddd DOLT

test_name	detail	row_cnt	sorted	mysql_time	sql_mult	cli_mult
batching	LOAD DATA	10000	1	0.07	1
batching	batch sql	10000	1	0.1	1.4
batching	by line sql	10000	1	0.09	1.67
blob	1 blob	200000	1	0.94	3.77	4.23
blob	2 blobs	200000	1	0.89	4.96	5.4
blob	no blob	200000	1	0.97	1.9	2.14
col type	datetime	200000	1	0.84	2.5	2.73
col type	varchar	200000	1	0.7	2.93	3
config width	2 cols	200000	1	0.77	2.51	2.31
config width	32 cols	200000	1	1.9	1.71	2.53
config width	8 cols	200000	1	0.99	2.09	2.32
pk type	float	200000	1	0.89	1.93	2.02
pk type	int	200000	1	0.81	2.09	2.2
pk type	varchar	200000	1	1.63	1.37	1.39
row count	1.6mm	1600000	1	5.73	2.44	2.53
row count	400k	400000	1	1.44	2.38	2.5
row count	800k	800000	1	2.88	2.39	2.49
secondary index	four index	200000	1	3.63	1.25	1.07
secondary index	no secondary	200000	1	0.92	2.01	2.17
secondary index	one index	200000	1	1.17	2.04	2.13
secondary index	two index	200000	1	2.05	1.53	1.47
sorting	shuffled 1mm	1000000	0	4.99	2.5	2.67
sorting	sorted 1mm	1000000	1	5.08	2.45	2.61

[github-actions]

@coffeegoddd DOLT

name	detail	mean_mult
dolt_blame_basic	system table	1.25
dolt_blame_commit_filter	system table	3.25
dolt_commit_ancestors_commit_filter	system table	0.84
dolt_commits_commit_filter	system table	0.91
dolt_diff_log_join_from_commit	system table	2.01
dolt_diff_log_join_to_commit	system table	2.06
dolt_diff_table_from_commit_filter	system table	1.13
dolt_diff_table_to_commit_filter	system table	1.14
dolt_diffs_commit_filter	system table	1
dolt_history_commit_filter	system table	1.37
dolt_log_commit_filter	system table	0.91

[github-actions]

@coffeegoddd DOLT

name	add_cnt	delete_cnt	update_cnt	latency
adds_only	60000	0	0	0.71
adds_updates_deletes	60000	60000	60000	3.74
deletes_only	0	60000	0	1.84
updates_only	0	0	60000	2.37