Truncate MCVs

[max-hoffman]

Sort and truncate MCVs. Only keep values whose frequency is > twice the uniform frequency. This prevents us from manually summing non-outliers (which is expensive).

[max-hoffman]

#benchmark

[github-actions]

@max-hoffman workflow run: https://github.com/dolthub/dolt/actions/runs/9602745041

[coffeegoddd]

@max-hoffman DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
0bc232f	ok	5937457

version	total_tests
0bc232f	5937457

correctness_percentage
100.0

[coffeegoddd]

@max-hoffman DOLT

test_name	from_latency_p95	to_latency_p95	is_faster
tpcc-scale-factor-1	74.46	74.46	0

test_name	server_name	server_version	tps	test_name	server_name	server_version	tps	is_faster
tpcc-scale-factor-1	dolt	b4dc473	33.54	tpcc-scale-factor-1	dolt	0bc232f	32.85	0

[coffeegoddd]

@coffeegoddd DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
39bcc83	ok	5937457

version	total_tests
39bcc83	5937457

correctness_percentage
100.0

[coffeegoddd]

@max-hoffman DOLT

read_tests	from_latency_median	to_latency_median	is_faster
covering_index_scan	2.81	2.81	0
groupby_scan	17.01	17.32	0
index_join	5.28	5.28	0
index_join_scan	3.25	2.52	1
index_scan	53.85	53.85	0
oltp_point_select	0.46	0.46	0
oltp_read_only	7.56	7.56	0
select_random_points	0.75	0.75	0
select_random_ranges	0.9	0.9	0
table_scan	54.83	54.83	0
types_table_scan	139.85	139.85	0

write_tests	from_latency_median	to_latency_median	is_faster
oltp_delete_insert	6.09	6.09	0
oltp_insert	3.02	3.02	0
oltp_read_write	13.95	13.95	0
oltp_update_index	3.07	3.07	0
oltp_update_non_index	3.02	3.02	0
oltp_write_only	6.43	6.43	0
types_delete_insert	6.67	6.67	0

[coffeegoddd]

@max-hoffman DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
c9875b3	ok	5937457

version	total_tests
c9875b3	5937457

correctness_percentage
100.0

[coffeegoddd]

@max-hoffman DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
ba8e7f4	ok	5937457

version	total_tests
ba8e7f4	5937457

correctness_percentage
100.0

[jycor]

LGTM

[nicktobey]

A bunch of tests now return no mcv information. I'd make sure that we're not losing any test coverage as a result. Otherwise, this looks good.

[nicktobey]

This should break out of the loop here, right?

[max-hoffman]

follow up bug fixes here #8049

[nicktobey]

Lets also add a test with multiple mcvs.

[nicktobey]

Why were these deleted?

[max-hoffman]

replaced with the analyze below, we already test background loading figured this was overkill