RFC - special case initial commit output

[Ian Clatworthy]

Robert Collins wrote:
> I was wondering, how about special casing the output of initial commit
> to default to quiet. This would
>  - address the overhead for folk benchmarking.
>  - not impact the regular case of commit showing what was altered
> 
> -Rob

I was wondering exactly the same thing last night. I'm very open to the
idea. Except for the initial commit case, I've kind of grown attached to
the verbosity of commit - seeing exactly what is being committed. On
initial commit, a count of items is arguably more useful if we wanted to
do that.

I suspect we'll receive a bit of a backlash on the "make verbose less
verbose" UI change if others feel the same way. So +1 from me.

Ian C.