More IR optimisation rounds

[xunilrj]

Description

Now that optimisation passes converge, we increase the number of rounds.

Checklist

• I have linked to any relevant issues.
• I have commented my code, particularly in hard-to-understand areas.
• I have updated the documentation where relevant (API docs, the reference, and the Sway book).
  • If my change requires substantial documentation changes, I have requested support from the DevRel team
• I have added tests that prove my fix is effective or that my feature works.
• I have added (or requested a maintainer to add) the necessary Breaking* or New Feature labels where relevant.
• I have done my best to ensure that my PR adheres to the Fuel Labs Code Review Standards.
• I have requested a review from the relevant team or maintainers.

[xunilrj]

e2e gas usage

	Improvements	Regressions
Count	1065	129
Average	0.53%	-2.26%
Median	0.17%	-2.16%
Max	54.51%	-8.04%
Min	0.01%	-0.01%

| should_pass/language/u256/u256_operators (test.toml)::should_be_able_to_use_comparison_operators | 510 | 232 | 54.51% |
| should_pass/language/intrinsics/storage_intrinsics (test.toml)::test_state_clear_slots_empty_slots | 1692 | 1828 | -8.04% |

e2e bytecode size

	Improvements	Regressions
Count	67	31
Average	4.73%	-17.02%
Median	1.71%	-4.09%
Max	57.14%	-154.97%
Min	0.05%	-0.21%

| should_pass/language/dereferenced_projection_reassignment (test.toml) | 168 | 72 | 57.14% |
| should_pass/language/references/impl_reference_types (test.toml) | 5632 | 14360 | -154.97% |

in lang gas usage

	Improvements	Regressions
Count	—	—
Average	—	—
Median	—	—
Max	—	—
Min	—	—

in lang bytecode size

	Improvements	Regressions
Count	—	—
Average	—	—
Median	—	—
Max	—	—
Min	—	—

o2 byte code

	Improvements	Regressions
Count	3	2
Average	1.39%	-1.53%
Median	1.47%	-1.53%
Max	1.61%	-2.09%
Min	1.10%	-0.97%

Test	Before	After	Percentage
./mocks/src20-token	19008	19192	-0.97%
./contracts/order-book	61144	62424	-2.09%
./contracts/order-book-registry	12312	12176	1.10%
./contracts/order-book-whitelist	5960	5864	1.61%
./contracts/order-book-blacklist	6000	5912	1.47%

o2 gas usage

	Improvements	Regressions
Count	—	1
Average	—	-0.22%
Median	—	-0.22%
Max	—	-0.22%
Min	—	-0.22%

Test	Before	After	Percentage
contract_libs::test_u56	463	464	-0.22%

[ironcev]

This is a bit surprising. Intuitively, I've expected mostly improvements and very few regressions. The bytecode size increase is likely coming from more inlining. Numerous increases in gas usage are not that easy to explain at the first look.

With this number of regressions, their unpredictably looking distribution, and negative impact on o2, I rather wouldn't proceed with this PR and stay at two rounds that brought good gains.

Or at least, I wouldn't proceed until we analyze the regressions in more detail and see the impact on o2 benchmarks. It would definitely be a good additional insight into our IR optimizations and pipeling to understand the regressions. But on the other side I am not sure if it is worth the effort and we just close this PR without merging and proceed with other optimizations where we know we will surely have good gains.