Use inbound peerslot slots when a substream is received, rather than a connection (#7464)

* Use inbound peerslot slots when a substream is received, rather than a connection

* Refactor PeerState

* Some bugfixes

* Fix warnings so that CI runs, gmlrlblbl

* Bugfixes

* Update docs

* Apply suggestions from code review

Co-authored-by: Roman Borschel <[email protected]>

* Clean up Banned state

* Refactor connections state

* Fix possibility of Enabled with no Opening or Open connection

* Line width

* Add some debug_asserts! and fix TODO

* Refactor legacy handler

* Rewrite group.rs entirely [part 1]

* Rewrite group.rs entirely [part 2]

* Remove faulty assertion

Because of the asynchronous nature of the behaviour <-> handler communications, it is possible to receive notifications while in the Closing state

* Don't poll the legacy substream is not Open

* Tolerate when not all substreams are accepted

* Remove TODOs

* Dummy commit to make CI log interesting things

* Try race condition fix

* Revert "Try race condition fix"

This reverts commit 0675c659.

* Correctly rebuild pending_opening

* Minor tweaks

* Printlns for CI debugging

* Revert "Printlns for CI debugging"

This reverts commit e7852a23.

* Revert "Dummy commit to make CI log interesting things"

This reverts commit 259ddd74

.

* mv group.rs ../handler.rs

* Apply suggestions from code review

Co-authored-by: Max Inden <[email protected]>

* Banned => Backoff

* Mention the actual PeerStates

* OpenDesired -> OpenDesiredByRemote

* OpeningThenClosing

* Add doc links to PeerState

* Simplify increment logic

* One more debug_assert

* debug_assert!

* OpenDesiredByRemote

* Update client/network/src/protocol/generic_proto/behaviour.rs

Co-authored-by: Max Inden <[email protected]>

Co-authored-by: Roman Borschel <[email protected]>
Co-authored-by: Max Inden <[email protected]>