Beautiful solution. Readable, concise and efficient.

On final submission are only the final tests run, or are the example tests run too?

Only final tests.

there is explicit guidance against being original in translations

There's a guidance to write good tests too. Anyway, it doesn't look too important due to the nature of the problem...

I don't disagree, but those cases are not in the C or Python translations, and there is explicit guidance against being original in translations.
On final submission are only the final tests run, or are the example tests run too?

I think the final tests should have fixed cases with 0 and 1 too; who knows what kind of wrong solutions can be made...

Thanks. Now fixed.

I agree. I seem to have copied that from the Python translation.

I've edited my Go translation to use a more naive implementation in the random tests.

Careful with references and order of execution... Now it can be solved with

func Fusc(n *big.Int) int {
  n.Set(new(big.Int))
  return 0
}

Made a Go translation. https://www.codewars.com/kumite/5c1906d75065a935ce000250

Made a Go translation. https://www.codewars.com/kumite/5c1680348a2d874d530002da

Updated with random test cases on submit.

that's good enough for me. i'll approve it

I see some kata have extra tests on submission, including randomly generated test cases.
I couldn't find a difference between "Test Cases" and "Sample Test Cases" for the exisiting languages in this kata, and the advice is not to be original.

Should there be a separate set of test that run when you submit?
I'm not sure what the difference between tests cases and sample test cases is supposed to be.

This is my first translation. Don't know if it's right to merge it myself without review.