optimize ForeignKey.HashOf()

[jycor]

This PR optimizes the ForeignKey.HashOf() function to be about 55% faster by using a sync.Pool of []byte instead of bytes.Buffer.

Unfortunately, this only accounts for like 0.2% of total tppc cpu runtime.

Old:

BenchmarkForeignKeyHashOf-14    	 2012511	       578.7 ns/op

New:

BenchmarkForeignKeyHashOf-14    	 4511584	       260.0 ns/op

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
e55f75f	ok	5937471

version	total_tests
e55f75f	5937471

correctness_percentage
100.0

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
aed3439	ok	5937471

version	total_tests
aed3439	5937471

correctness_percentage
100.0

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
691bd70	ok	5937471

version	total_tests
691bd70	5937471

correctness_percentage
100.0

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
b14dfed	ok	5937471

version	total_tests
b14dfed	5937471

correctness_percentage
100.0

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
7b8b0ef	ok	5937471

version	total_tests
7b8b0ef	5937471

correctness_percentage
100.0

[jycor]

#benchmark

[github-actions]

@jycor workflow run: https://github.com/dolthub/dolt/actions/runs/22420871058

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
9c4bcac	ok	5937471

version	total_tests
9c4bcac	5937471

correctness_percentage
100.0

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
c0932c5	ok	5937471

version	total_tests
c0932c5	5937471

correctness_percentage
100.0

[angelamayxie]

Is there a way to avoid using append?

[jycor]

It's probably possible to rewrite this without using append(), but I don't think it would be any different performance-wise?
If my understanding is correct, append() should only cause reallocations when the new entry exceeds capacity.

[zachmu]

LGTM

[coffeegoddd]

@jycor DOLT

comparing_percentages
100.000000 to 100.000000

version	result	total
e5a6ba7	ok	5937471

version	total_tests
e5a6ba7	5937471

correctness_percentage
100.0

[coffeegoddd]

@jycor DOLT

test_name	from_latency_p95	to_latency_p95	percent_change
tpcc-scale-factor-1	61.08	61.08	0.0

test_name	from_server_name	from_server_version	from_tps	to_server_name	to_server_version	to_tps	percent_change
tpcc-scale-factor-1	dolt	448f95c	37.85	dolt	e5a6ba7	38.03	0.48