feat(perf): Simplify array get when most recent array set writes to the same index

[vezenovm]

Description

Problem*

Part of general effort to reduce Brillig bytecode sizes

Summary*

This takes inspiration from #6207 that the value is still known at a previous array set when the index is dynamic.

Additional Context

Documentation*

Check one:

• No documentation needed.
• Documentation included in this PR.
• [For Experimental Features] Documentation to be submitted in a separate PR.

PR Checklist*

• I have tested the changes locally.
• I have formatted the changes with Prettier and/or cargo fmt on default settings.

[github-actions]

Changes to Brillig bytecode sizes

Generated at commit: 7cc3c5bf5b266e459a070fd49f5afe60cdc4b682, compared to commit: c3cb38a7c4de6fc321b367eda3fca6d06e76b77a

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
slice_dynamic_index	-28 ✅	-1.09%
nested_dyn_array_regression_5782	-10 ✅	-7.87%

---

Full diff report 👇

Program	Brillig opcodes (+/-)	%
slice_dynamic_index	2,547 (-28)	-1.09%
nested_dyn_array_regression_5782	117 (-10)	-7.87%

[github-actions]

Changes to circuit sizes

Generated at commit: 7cc3c5bf5b266e459a070fd49f5afe60cdc4b682, compared to commit: c3cb38a7c4de6fc321b367eda3fca6d06e76b77a

🧾 Summary (10% most significant diffs)

Program	ACIR opcodes (+/-)	%	Circuit size (+/-)	%
regression_mem_op_predicate	-6 ✅	-10.34%	-11 ✅	-0.31%
slice_dynamic_index	-18 ✅	-1.50%	-35 ✅	-0.55%

---

Full diff report 👇

Program	ACIR opcodes (+/-)	%	Circuit size (+/-)	%
nested_dyn_array_regression_5782	30 (-2)	-6.25%	2,832 (-5)	-0.18%
regression_mem_op_predicate	52 (-6)	-10.34%	3,552 (-11)	-0.31%
slice_dynamic_index	1,182 (-18)	-1.50%	6,384 (-35)	-0.55%

[TomAFrench]

I don't think this takes into account enable_side_effects. Can you add a test for this?

fn main(v1: [Field; 3], v2: u32) {
   enable_side_effects u1 0
   v4 = array_set v1, index v2, value v3
   enable_side_effects u1 1
   v5 = array_get v4, index v2
   return v5
}

[vezenovm]

I don't think this takes into account enable_side_effects

Ah yes good catch. I'm not sure if we can have this optimization here then as we need more information than just the array instructions. Going to convert this back to a draft for now.

[vezenovm]

Closing in as we need to do this during constant folding to take into account the predicate #6684