chore: adding benchmarking tools for number of brillig opcodes

[kashbrti]

Description

Problem*

Resolves

Summary*

Additional Context

PR Checklist*

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

[TomAFrench]

Looks like you're not making use of --force-brillig here which would simplify things a lot.

[TomAFrench]

See https://github.com/noir-lang/noir/blob/master/test_programs/gates_report_brillig.sh for some inspiration (although you can't make use of nargo info.)

[kashbrti]

should I do the --force-brillig when doing the nargo export?

[TomAFrench]

Yeah you need to specify that you want brillig output when you compile.

[kashbrti]

@TomAFrench I'm wondering if the format of the outputs generated by the script is compatible with the noir_gate_diff action. I'm not finding documentation explaining the json format required by the gate_diff

[TomAFrench]

https://github.com/noir-lang/noir-gates-diff/blob/dbe920a8dcc3370af4be4f702ca9cef29317bec1/src/types.ts#L9-L12

The action expects a JSON object {"name": string, "opcodes": number} for each brillig function we're tracking.

[TomAFrench]

Ah, I see you changed to that. Will take a closer look.

[kashbrti]

ah then it should be good. The objects in the Json are called unconstrained_function instead of function though. should I change that or is that expected?

[TomAFrench]

That's expected.

[TomAFrench]

This needs to be marked as a brillig report

[TomAFrench]

https://github.com/noir-lang/noir/blob/0a64e8ee6497ef228a92f42af0bdb5081025bf37/.github/workflows/reports.yml#L135