Chore: Update version 0.5.0

[jcabrero]

This PR includes the changes required to update Nada-Numpy to version 0.5.0 corresponding to version 0.6.0 of Nada-DSL and Nillion Python Client.

Note: This is expected to fail before finishing we need to:

• Version 0.6.0 of Nada DSL and Python Client needs to be released.
• Update poetry lock: poetry lock
• Push to and wait for CI/CD to execute.

[mathias-nillion]

Looks good once CI/CD passes - left some comments/ideas but I think they fall outside of the scope of this PR

[mathias-nillion]

interesting change. how do we know if these values are public/secret and signed/unsigned?
can we no longer customize this is these YAML files?

[jcabrero]

Basically, the value is introduced into the specific type expected by the Nada program. So it takes it already from the Nada Program.

[mathias-nillion]

If we can no longer type-annotate in these YAMLs, that removes any prospect of annotating as Rational to abstract away this scaling logic?
Or do we just assume that if a float is passed then it is a rational?

[mathias-nillion]

Not a big fan of this :p
Same goes for all other instances of this.

[jcabrero]

In this new version, programs without compute stop compiling. All these programs where there's this instance of a += Integer(0) are programs that without this compute line would not compile. I know it's a tricky thing, but I don't know a better/faster way to handle it.

[mathias-nillion]

Looking at this, it would actually be quite nice if we could put a (possibly nested) list in these YAMLs
So something like A: [[1, 2, 3], [4, 5, 6], ...]

[jcabrero]

WIth the new testing infrastructure, I would like to deprecate the use of these files and creating programmatic tests that check for all these behaviours. In such case, we can directly input lists and forget about filling up ginormous files.