In practice we always create a remote and connect to pipewire.
Any other scenario is invalid, therefore, it is not justified
to be confused with so many classes for such small functionality.
This simplifies a lot the modules code.

Also, this commit exposes the pw_core and pw_remote objects
out of WpCore. This is in practice useful when dealing with low-level
pw and spa factories, which are used in the monitors. Let's not
add API wrappers for everything... Bindings will never use this
functionality anyway, since it depends on low level pipewire C API.

+ add a method to retrieve the core
+ change the register function to not require the core as a parameter

This is a cleaner way to interface with the remote pipewire daemon.
The WpRemote base class can be subclassed also for interfacing
with other daemons (hardware-specific managers, etc)

We need to delay the registration of the simple endpoint because
it needs to get some information from the "info" callback of pipewire
before we can use it in a link.

This is implemented in a slightly hacky way, we register the GMainLoop
as a global object and use it from the module to quit the daemon.
This is bad design because the module assumes it is loaded inside
our daemon.

In the future, this should change. It looks like we should have an
object that tracks the state of PipeWire and main() should track
state changes of that object and decide what to do.

This changes the registration point to be the endpoint and not the
session manager, as it seems easier to implement (the implementation
will not need to keep a pointer to the session manager around)

After discussing things at the AGL May 2019 F2F meeting
and reflecting on the initial design of WirePlumber,
it became clear that it needed a fresh start.