[RFA] fix *stopped for CLI commands

[Daniel Jacobowitz]

On Fri, Feb 06, 2009 at 10:45:18AM +0300, Vladimir Prus wrote:
> This patch fixes this by making MI observer print frame again, into MI uiout,
> if necessary. It passes all MI tests in (sync,async)x(native,gdbserver) combinations.
> How does this look, and are non-MI changes OK? If approved, I'll add a test that
> CLI commands result in proper *stopped.

Aside from the docs issue, this looks good to me.

-- 
Daniel Jacobowitz
CodeSourcery