fix: log startup errors to stderr and exit non-zero

When start() returns an error, main() logged it via the slog logger and
returned normally (exit 0), so a crash-looping pod showed as "Completed"
with no error in the logs: start() defers the OTel SDK shutdown, tearing
down the log exporter before main() logs, so with LOG_FORMAT=otel the
record never reaches Alloy.

Also write the error to stderr (always captured by `kubectl logs`) and
os.Exit(1) so the container is correctly reported as failed.

cmd/service/service.go

@@ -66,6 +66,13 @@ func main() {
 		cli,
 	); err != nil {
 		logger.With("error", err).Error("process error")
+		// start() defers the OTel SDK shutdown, so by the time we get here the
+		// log exporter may already be torn down and the line above never
+		// reaches Alloy. Write to stderr too so startup failures are always
+		// visible in `kubectl logs`, and exit non-zero so the container is
+		// reported as failed (CrashLoopBackOff) instead of "Completed".
+		fmt.Fprintf(os.Stderr, "fatal: process error: %v\n", err)
+		os.Exit(1)
 	}
 }