feat: add (x | 1) optimization for booleans

[jfecher]

Description

Problem*

Resolves an issue in a slack thread by @michaeljklein

Summary*

Breakout of a change in #6771 where slice pop back seemed to be using all of its ctrl_typevars for a tuple instead of the one element that was actually being popped. This only affects the ArrayGet though which is subsequently optimized out.

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 number of Brillig opcodes executed

Generated at commit: 9f0fab59e72cb848619959527679ebf75db780c1, compared to commit: 919149d3413be5232b33611094687fdb5fd86086

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
bool_or	-2 ✅	-8.00%

---

Full diff report 👇

Program	Brillig opcodes (+/-)	%
bool_or	23 (-2)	-8.00%

[github-actions]

Changes to Brillig bytecode sizes

Generated at commit: 9f0fab59e72cb848619959527679ebf75db780c1, compared to commit: 919149d3413be5232b33611094687fdb5fd86086

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
bool_or	-4 ✅	-12.90%

---

Full diff report 👇

Program	Brillig opcodes (+/-)	%
bool_or	27 (-4)	-12.90%

[github-actions]

Peak Memory Sample

Program	Peak Memory
keccak256	79.14M
workspace	121.92M
regression_4709	295.98M
ram_blowup_regression	2.44G
private-kernel-tail	210.43M
private-kernel-reset	862.21M
private-kernel-inner	307.94M
parity-root	175.78M

[github-actions]

Compilation Sample

Program	Compilation Time	%
sha256_regression	0m1.528s	5%
regression_4709	0m0.832s	9%
ram_blowup_regression	0m17.393s	1%
private-kernel-tail	0m1.154s	-7%
private-kernel-reset	0m9.373s	3%
private-kernel-inner	0m2.803s	3%
parity-root	0m0.911s	-4%
noir-contracts	2m58.779s	9%

[michaeljklein]

LGTM

[jfecher]

$ cargo t partial_constant_folding
   Compiling noirc_evaluator v1.0.0-beta.0 (/home/jake/Code/Noir/noir/compiler/noirc_evaluator)
    Finished test [optimized + debuginfo] target(s) in 1.09s
     Running unittests src/lib.rs (/home/jake/Code/Noir/noir/target/debug/deps/noirc_evaluator-4693fae089867dbc)

running 0 tests

test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 189 filtered out; finished in 0.00s

I must be missing something obvious. Not sure why partial_constant_folding is failing in CI but not being run at all locally for me.

Edit: Maybe it is the #[cfg(feature = "bn254")]