fix login throttle handling

We were checking the login throttle at the beginning of every SASL
conversation. This had several problems:

1. Pidgin (on Windows?) tries every mechanism in order, regardless of
the CAP advertisement. It would use up the default throttle allowance
trying unsupported mechanisms like CRAM-MD5.
2. The throttle was actually checked twice for AUTHENTICATE PLAIN
(once at the start of the conversation and once in AuthenticateByPassphrase).

The general pattern here is that we should check the throttle every time we
do something "expensive" (bcrypt verification, send a reset email) or
"dangerous" (anything that could lead to a bruteforce attack on passwords).
Therefore, delete the check from the AUTHENTICATE handler, and add one at
the beginning of the SCRAM conversation to replace it.

irc/handlers.go

@@ -207,13 +207,6 @@ func authenticateHandler(server *Server, client *Client, msg ircmsg.Message, rb
 
 	// start new sasl session: parameter is the authentication mechanism
 	if session.sasl.mechanism == "" {
-		throttled, remainingTime := client.loginThrottle.Touch()
-		if throttled {
-			rb.Add(nil, server.name, ERR_SASLFAIL, client.Nick(),
-				fmt.Sprintf(client.t("Please wait at least %v and try again"), remainingTime.Round(time.Millisecond)))
-			return false
-		}
-
 		mechanism := strings.ToUpper(msg.Params[0])
 		_, mechanismIsEnabled := EnabledSaslMechanisms[mechanism]
 
@@ -384,6 +377,12 @@ func authScramHandler(server *Server, client *Client, session *Session, value []
 
 	// first message? if so, initialize the SCRAM conversation
 	if session.sasl.scramConv == nil {
+		if throttled, remainingTime := client.checkLoginThrottle(); throttled {
+			rb.Add(nil, server.name, ERR_SASLFAIL, client.Nick(),
+				fmt.Sprintf(client.t("Please wait at least %v and try again"), remainingTime.Round(time.Millisecond)))
+			continueAuth = false
+			return false
+		}
 		session.sasl.scramConv = server.accounts.NewScramConversation()
 	}