Questionable undocumented inout fix

[Timon Gehr via Digitalmars-d]

On 09/13/2014 07:54 PM, Kagamin wrote:
>  From https://issues.dlang.org/show_bug.cgi?id=10850:
>> Since apparently inout now refers to the outermost inout function
>
> This sounds questionable. Anybody knows, what happened?

It's one of Kenji Hara's occasional stealth language 
changes/refinements. It was pulled because it fixed a certain kind of 
type system unsoundness, but I don't remember which issue/pull this was.