chore(fuzz): Compare print output on failed programs

[aakoshh]

Description

Problem*

Follow up for a discussion about whether we can treat failures as equivalents.

Summary*

So far the fuzzer considered the programs to either succeed with an optional return value and the printed output, or fail with an error. Now the printed output is also considered on the failure side: if the errors are considered equivalent, then we also demand that the printed output is the same, for the result to be "both failed the same way, therefore the test passes".

Additional Context

I didn't go all the say to say all ACIR and Brillig errors are equivalent just yet. Let's relax them as they crop up, knowing that the backstop of comparing the prints is there.

I relaxed the ConstantDoesNotFitType error in the SSA interpreter to be considered equivalent if the constant is the same and the type is the same. I didn't do it for other errors for now, because the SSA interpreter does not capture print outputs yet.

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]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Execution Time'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.

Benchmark suite	Current: a603ece	Previous: 2e36b30	Ratio
rollup-merge	0.004 s	0.003 s	1.33

This comment was automatically generated by workflow using github-action-benchmark.

CC: @TomAFrench