Add method to use the ENCODE-OBJECT/ENCODE-SLOTS protocol in ENCODE

[fiddlerwoaroof]

Implements #66.

The current state of the library makes the incremental encoder a completely different world from ENCODE.
This makes it harder to extend YASON's encoding system because you have to override both ENCODE and
ENCODE-SLOTS.

[phmarek]

That's a bit too unspecific... Won't work for ratios, complex numbers, and so on.

How about adding a STANDARD-OBJECT qualification? Does that code also work for structures?
Please include a test as well.

Thanks for the idea!

[phmarek]

Did you mean to change anything? I can't see any difference in the diff.

Can you add an example or at least a test as well, please?

Thanks!