#14259: fix(discord): add timeout to restart to prevent duplicate responses

Restarts by openclaw were too quick to restart, resulting in more than one bot per channel, and duplicate responses. <!-- greptile_comment --> <h2>Greptile Overview</h2> <h3>Greptile Summary</h3> Adds a systemd service example file (`docs/install/systemd.service.example`) to prevent duplicate Discord bot responses caused by overly fast gateway restarts. The key fix is `RestartSec=10`, which introduces a 10-second delay between restart attempts, giving the previous bot instance time to fully disconnect before a new one starts. - Adds new systemd unit file at `docs/install/systemd.service.example` with `RestartSec=10` to throttle restarts - Configures the gateway to run as a dedicated `openclaw` user under `/var/lib/openclaw` - Note: existing docs reference `systemd --user` services, while this example uses a system-level unit — both are valid deployment models but this difference is worth documenting <h3>Confidence Score: 4/5</h3> - This PR is safe to merge — it only adds a documentation example file with no code changes. - The change is a single new example file in the docs directory with no runtime code impact. The systemd configuration follows standard conventions and the `RestartSec=10` value is reasonable for preventing duplicate bot instances. Minor note: the system-level service pattern differs from the existing user-level systemd references in the docs, but both are valid approaches. - No files require special attention. <!-- greptile_other_comments_section --> _{(2/5) Greptile learns from your feedback when you react with thumbs up/down!} <!-- /greptile_comment -->