dce/rpc "client" api

[Andrew Tridgell]

> it is also _not_ a new protocol.  it's a transport. it does absolutely
> _nothing_ - zero modifications - to the Dce/rpc data it transports.  it
> therefore has no effect on any code reviews required of the marshalling /
> unmarshalling, zero effect on any code reviews required of the dce/rpc
> service implementations.

There is a bit more to communicating with pipe daemons than just
passing along pdu data. You also need to pass along (and potentially
pass back) state information about smbd. The involves some sort of
protocol. The most obvious place where you have met this already is
the authentication handling.

We are migrating most smbd state information into databases so that it
is accessible between processes, but that migration is nowhere near
complete. It is also not clear that we actually want all variables to
be in a database as the access cost does go up quite a lot.