fix: reset reconnect backoff after successful connection

CRITICAL: The agent never reset its exponential backoff delay after
successful connections, causing permanent 60s reconnect delays after
multiple deploys.

Root cause:
- retryDelay started at 1s, doubled to 60s cap on each disconnect
- Never reset when connection succeeded
- After ~6 deploys, agent permanently waited 60s between reconnects

Fixes:
1. Reset backoff to 1s after session runs for 30s+ (indicates stable connection)
2. Reduce max retry from 60s → 10s (60s is too long for production)

Impact:
- After Phoenix deploy, agent reconnects within seconds instead of minutes
- Exponential backoff still prevents connection storms on real failures
- Typical reconnect sequence: 1s → 2s → 4s → 8s → 10s (cap)

agent.go

@@ -42,7 +42,8 @@ type channelMsg struct {
 func runAgent(ctx context.Context, wsURL, token string) {
 	baseURL := strings.TrimRight(wsURL, "/")
 	retryDelay := time.Second
-	maxRetry := 60 * time.Second
+	maxRetry := 10 * time.Second
+	const successfulConnectionThreshold = 30 * time.Second
 
 	for {
 		select {
@@ -51,7 +52,18 @@ func runAgent(ctx context.Context, wsURL, token string) {
 		default:
 		}
 
+		sessionStart := time.Now()
 		err := runSession(ctx, baseURL, token)
+		sessionDuration := time.Since(sessionStart)
+
+		// Reset backoff if session ran successfully for a while (indicates stable connection)
+		if sessionDuration >= successfulConnectionThreshold {
+			slog.Debug("resetting reconnect backoff after successful session",
+				"duration", sessionDuration,
+				"previous_delay", retryDelay)
+			retryDelay = time.Second
+		}
+
 		if ctx.Err() != nil {
 			return
 		}