resolvedOptions ordering

[Constellation]

Looking into the spec of DurationFormat, and one thing I found is that the ordering of resolvedOptions is inconsistent to resolvedOptions in RelativeTimeFormat.
https://tc39.es/ecma402/#sec-intl.relativetimeformat.prototype.resolvedoptions
In RelativeTimeFormat,

1. locale
2. style
3. numeric
4. numberingSystem

In DurationFormat,

1. locale
2. numberingSystem
3. style
4. fields

While processing timings of numberingSystem and style are similar in DurationFormat and RelativeTimeFormat, their orderings are different. Currently I'm not sure which one should be appropriate, but I think we should make them consistent. Is there a policy about it?

[Constellation]

/CC @rkirsling

[FrankYFTang]

This is NOT a editorial change but a normative change. While I am not not in particular think the consistency is important, the change of order will be a normative observable changes. I think you need to take out the "editorial' label.

[ryzokuken]

@Constellation as Frank said, I don't care either way either. I will make this consistent once #32 is resolved.

[ryzokuken]

This is fixed in the latest update. Closing.