Testing: Compile twice and hash bytecode to check for non-determinism

[vezenovm]

Problem

We should have logic as part of our test-suite to make sure that we don't produce non-deterministic circuits for the same Noir code. This fix (#2035) was recently pushed through as 9_conditional was generating non-deterministic bytecode.

Happy Case

We should catch any non-determinism in our test-suite.

Alternatives Considered

Re-running compilation twice will slow down our CI. Perhaps we should note any critical areas where there is a higher potential for non-determinism (such as merging conditional branches) and only re-compile these programs.

Additional Context

No response

Would you like to submit a PR for this Issue?

No

Support Needs

No response