feat(swarm): make stream uprade errors more ergonomic

The currently provided `ConnectionHandlerUpgrErr` is very hard to use. Not only does it have a long name, it also features 3 levels of nesting which results in a lot of boilerplate. Last but not least, it exposes `multistream-select` as a dependency to all protocols.

We fix all of the above by renaming the type to `StreamUpgradeError` and flattening out its interface. Unrecoverable errors during protocol selection are hidden within the `Io` variant.

Related: #3759.

Pull-Request: #3882.

protocols/ping/src/handler.rs

@@ -23,13 +23,12 @@ use futures::future::BoxFuture;
 use futures::prelude::*;
 use futures_timer::Delay;
 use libp2p_core::upgrade::ReadyUpgrade;
-use libp2p_core::{upgrade::NegotiationError, UpgradeError};
 use libp2p_swarm::handler::{
     ConnectionEvent, DialUpgradeError, FullyNegotiatedInbound, FullyNegotiatedOutbound,
 };
 use libp2p_swarm::{
-    ConnectionHandler, ConnectionHandlerEvent, ConnectionHandlerUpgrErr, KeepAlive,
-    NegotiatedSubstream, StreamProtocol, SubstreamProtocol,
+    ConnectionHandler, ConnectionHandlerEvent, KeepAlive, NegotiatedSubstream, StreamProtocol,
+    StreamUpgradeError, SubstreamProtocol,
 };
 use std::collections::VecDeque;
 use std::{
@@ -238,14 +237,14 @@ impl Handler {
         self.outbound = None; // Request a new substream on the next `poll`.
 
         let error = match error {
-            ConnectionHandlerUpgrErr::Upgrade(UpgradeError::Select(NegotiationError::Failed)) => {
+            StreamUpgradeError::NegotiationFailed => {
                 debug_assert_eq!(self.state, State::Active);
 
                 self.state = State::Inactive { reported: false };
                 return;
             }
             // Note: This timeout only covers protocol negotiation.
-            ConnectionHandlerUpgrErr::Timeout => Failure::Timeout,
+            StreamUpgradeError::Timeout => Failure::Timeout,
             e => Failure::Other { error: Box::new(e) },
         };