Implement SpacetimeType for Result<T, E>

[mamcx]

Description of Changes

Closes #3673

NOTE: C++ part will be in another PR

Expected complexity level and risk

2

Adding a new type touch everywhere

Testing

• Adding smoke and unit test

[JasonAtClockwork]

I'll build out an attached PR to add the new functionality for the Unreal SDK.

[JasonAtClockwork]

There is updates to the /modules/sdk-test-cs but this won't be enough testing for the C# side. We'll need to add tests into the /sdks/csharp/examples~/regression-tests like for UUID to make sure this works end-to-end.

@rekhoff - I'll also want your eyes on this one when you can for the C# side

[JasonAtClockwork]

I'm guessing this was added as the Result has two types and that was giving trouble with the BSATNName property directly?

Two notes on this, this should be ToBSATNString() or something like it to match PascalCase for the C# convention followed in all the other code. Second BSATNName was heavily used in CodeGen/Module.cs and doesn't seem to be updated in all locations. This will need review to make sure it won't cause problems for Result<> types used throughout the codegen.

[mamcx]

Yes, for things like:

SpacetimeDB.Result<SpacetimeDB.Sdk.Test.Module.EveryPrimitiveStruct, string>.BSATN<SpacetimeDB.Sdk.Test.Module.EveryPrimitiveStruct, string, SpacetimeDB.Sdk.Test.Module.EveryPrimitiveStruct.BSATN, SpacetimeDB.BSATN.String> rRW = new();

But is not possible to just pass the BSATNName everywhere, I need to manually pick some places.

[gefjon]

I'm happy with the Rust parts of this PR. We'll still need review from the other languages' maintainers.

[rekhoff]

This will confuse the XML parser. We should use the < and > versions of the characters in summary blocks.

Suggested change

	/// A use of a Result<T, E> type.
	/// A use of a Result&lt;T, E&gt; type.

[rekhoff]

The C# side looks good. I did some local testing of end-to-end behavior, and things look good.

[coolreader18]

Seems reasonable to me, except for my one comment.

[coolreader18]

Shouldn't this be -> Result<Result<AlgebraicType, AlgebraicType>, Self>?

[mamcx]

Mm, this is like the option variant:

/// Converts `self` into an `Option<AlgebraicValue>`, if applicable.
    pub fn into_option(self) -> Result<Option<Self>, Self> {
        match self {
            AlgebraicValue::Sum(sum_value) => match sum_value.tag {
                0 => Ok(Some(*sum_value.value)),
                1 => Ok(None),
                _ => Err(AlgebraicValue::Sum(sum_value)),
            },
            _ => Err(self),
        }
    }