perf: adding rollup cache for massive improvement of builds

[diervo]

Details

Allow rollup to cache per project (or rollup configuration)
This allow for a massive improvement while building

Does this PR introduce a breaking change?

• Yes
• No

[salesforce-best-lwc-internal]

Benchmark results

Base commit: ac254c5 | Target commit: 3e1c754

simple-benchmark-prod

object_keys	metric	base(ac254c5)	target(3e1c754)	trend
stringify_parse	duration	35.80 (±0.30 ms)	35.50 (±0.20 ms)	-0.3ms (0.8%) 👍
deep_merge	duration	28.80 (±0.60 ms)	28.00 (±0.60 ms)	-0.8ms (2.8%) 👍

simple-benchmark-compat

object_keys	metric	base(ac254c5)	target(3e1c754)	trend
stringify_parse	duration	36.10 (±0.80 ms)	35.00 (±0.30 ms)	-1.1ms (3.0%) 👍
deep_merge	duration	105.30 (±3.90 ms)	99.70 (±2.20 ms)	-5.6ms (5.3%) 👍

lwc-examples-prod

simple-item	metric	base(ac254c5)	target(3e1c754)	trend
create_and_render	duration	53.70 (±1.50 ms)	53.10 (±2.10 ms)	-0.6ms (1.1%) 👌

lwc-examples-compat

simple-item	metric	base(ac254c5)	target(3e1c754)	trend
create_and_render	duration	96.90 (±4.35 ms)	165.20 (±9.60 ms)	+68.3ms (70.5%) 👎

[salesforce-best-lwc-internal]

Benchmark results

Base commit: ac254c5 | Target commit: 0c60943

simple-benchmark-prod

object_keys	metric	base(ac254c5)	target(0c60943)	trend
stringify_parse	duration	35.80 (±0.30 ms)	35.50 (±0.40 ms)	-0.3ms (0.8%) 👍
deep_merge	duration	28.80 (±0.60 ms)	28.70 (±0.70 ms)	-0.1ms (0.3%) 👌

simple-benchmark-compat

object_keys	metric	base(ac254c5)	target(0c60943)	trend
stringify_parse	duration	36.10 (±0.80 ms)	35.70 (±0.40 ms)	-0.4ms (1.1%) 👍
deep_merge	duration	105.30 (±3.90 ms)	105.90 (±2.20 ms)	+0.6ms (0.6%) 👌

lwc-examples-prod

simple-item	metric	base(ac254c5)	target(0c60943)	trend
create_and_render	duration	53.70 (±1.50 ms)	53.70 (±2.50 ms)	0.0ms (0.0%) 👌

lwc-examples-compat

simple-item	metric	base(ac254c5)	target(0c60943)	trend
create_and_render	duration	96.90 (±4.35 ms)	95.15 (±3.35 ms)	-1.8ms (1.8%) 👌