Remove MOVE $$locbase, $$sp if non-entry function doesn't use stack

[ironcev]

Description

This PR:

• removes redundant MOVE $$locbase, $$sp if non-entry function doesn't use stack at all.
• fixes Register pressure makes compiler panic in register_allocator::spill #7547

If the function does not use stack at all, the $$locbase is never used, by definition.

Not using stack at all means that the function:

• does not have any locals
• does not call any functions that spill arguments to the stack
• does not spill any registers to the stack when allocating registers

Having MOVE $$locbase, $$sp in entry functions is a part of compiler's contract and tooling relies on it. Therefore, MOVE $$locbase, $$sp remains in entry functions even if $$locbase is not used.

On the contrary, in non-entry functions, setting $$locbase is an implementation detail no tooling relies on. E.g. a forc debug or similar.

Additionally, the PR:

• groups function prologue and epilogue instructions into [fn/entry init/end] groups for better readability and easier analysis of ASM. Also, it opens the way towards better automatic analysis of ASM output if needed in the future.

• adapts filter-fn to use the new [fn/entry init/end: <fn name>]comments.

• Fixes Register pressure makes compiler panic in register_allocator::spill #7547.

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.

[ironcev]

👍

[codspeed-hq]

Merging this PR will degrade performance by 20.27%

❌ 1 regressed benchmark
✅ 24 untouched benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
❌	tokens_for_program	522.6 µs	655.5 µs	-20.27%

---

_{Comparing ironcev/remove-setting-locbase (f8651cf) with master (c6d0d12)}

[cursor]

PR Summary

Medium Risk
Touches core Fuel codegen and register-allocation/optimization pipelines; while behavior should be equivalent, incorrect stack-usage detection or metadata plumbing could change generated bytecode and tooling assumptions.

Overview
Reduces generated Fuel VM bytecode by dropping redundant function prologue setup: MOVE $$locbase, $$sp is now removed for non-entry functions proven to not use the stack (detected via CFEI 0), while keeping the instruction for entry functions to preserve tooling expectations.

To support this and improve debuggability, instruction sets now carry optional (function name, is_entry) metadata through register allocation/optimization, and emitted ASM comments are standardized into [entry|fn init/end: <name>] groups (calls and returns included). This refactors the Fuel register allocator API to return Vec<AllocatedAbstractOp>, adds Value::get_parent_function(), updates module accessors, and refreshes affected e2e snapshots plus a new no_locbase optimization test.

^{Written by Cursor Bugbot for commit f8651cf. This will update automatically on new commits. Configure here.}

[ironcev]

Comparing 'e2e-gas-usages':

	Improvements	Regressions
Count	268	—
Average	1.65%	—
Median	1.61%	—
Max	8.22%	—
Min	0.01%	—

Comparing 'e2e-bytecode-sizes':

	Improvements	Regressions
Count	188	—
Average	5.09%	—
Median	2.86%	—
Max	12.50%	—
Min	0.03%	—

Comparing 'in-language-gas-usages':

	Improvements	Regressions
Count	418	—
Average	1.43%	—
Median	1.37%	—
Max	4.44%	—
Min	0.01%	—

Comparing 'bytecode-sizes' for project 'o2':

	Improvements	Regressions
Count	12	—
Average	0.14%	—
Median	0.12%	—
Max	0.21%	—
Min	0.09%	—

Comparing 'gas-usages' for project 'o2':

	Improvements	Regressions
Count	65	—
Average	0.06%	—
Median	0.04%	—
Max	1.69%	—
Min	0.02%	—

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.}