[intrinsics] moving the BSMs to java.util.Formatter

[Brian Goetz]

Rationale: The BSM for intrinsifying a given library behavior — and which is bound by the spec of that behavior — is logically part of that library, and should be co-maintained with that library.  

Sent from my iPad

> On Apr 8, 2019, at 10:53 PM, Vicente Romero <vicente.romero at oracle.com> wrote:
> 
> Hi,
> 
> Today in an offline chat, Brian suggested that the BSM for the intrinsification of String::format should be moved to java.util.Formatter. The BSM is accessing a lot of private methods in j.u.Formatter so it makes more sense to be there plus this will allow the BSM plus the library code to co-evolve.
> 
> Comments?
> Vicente