feat: SSA interpreter checks return value

[asterite]

Description

Problem

No issue, but I think @aakoshh mentioned this as a potential improvements.

This will also be useful to run nargo interpret on all execution_success test programs, combined with #8882.

Summary

For example, given this program:

fn main() -> pub Field {
    0
}

and this Prover.toml:

return = 1

running nargo interpret we get this output:

--- Interpreter result after Initial SSA:
Ok([Numeric(Field(0))])
---
Error: interpreter produced an unexpected result.
Expected result: [Numeric(Field(1))]
Actual result:   [Numeric(Field(0))]

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.

[aakoshh]

Great! Not sure if it has to be a shortcutting issue, but I suppose if we know it's wrong it's unlikely to get back on track in subsequent passes.

[aakoshh]

Although actually one of the bugs you fixed did correct itself, so we would have to carry the “any disagreed” flag to the end.

Never mind though, I’m happy either way.

[asterite]

I think that even if the return value changes in an intermediate pass and then changes again later on to the correct value, it might mean there's a bug in the interpreter so maybe the value fixing itself is just a coincidence.