dnstt fails with default record config

Start by locating the dnstt service template, such as run.sh.j2, and inspect how an empty record_type reaches the ExecStart line in /other/dnstt. Reproduce with an empty record_type and check the generated unit using journalctl -u hiddify-dnstt-.service. Done means the service no longer emits an argumentless -record-type flag and dnstt-server starts successfully.

[BUG] dnstt-server fails to start (crash loop) when record_type field is left empty in the panel

Description:
When adding or editing a dnstt domain via the Hiddify-Manager panel, leaving the record_type field empty results in a broken systemd service file. Instead of defaulting to txt or omitting the flag entirely, the backend generates the -record-type flag without any argument. This causes the dnstt-server to crash immediately with a 2/INVALIDARGUMENT status.

Steps to Reproduce:

Go to the Hiddify panel and configure a domain with dnstt transport.

Leave the record_type advanced field empty ("") (as is)

Apply configurations.

Check the status of the generated service using journalctl -u hiddify-dnstt-<domain>.service (the one in /other/dnstt)

Expected Behavior:
If the field is empty, the templating engine (e.g., run.sh.j2) should either fallback to "txt" or omit the -record-type flag from the ExecStart line so the dnstt-server binary can use its own default value.

Actual Behavior:
The systemd service file is generated with an incomplete flag at the end of the ExecStart line:
Ini, TOML

ExecStart=/opt/hiddify-manager/other/dnstt/dnstt-server ... -keepalive 0s -idle-timeout 0s -record-type 

This causes the daemon to crash.

Logs / Evidence:

May 29 19:20:15 server-1xgjqp systemd[1]: hiddify-dnstt-t.example.com.service: Main process exited, code=exited, status=2/INVALIDARGUMENT
May 29 19:20:20 server-1xgjqp dnstt-server[1080423]: flag needs an argument: -record-type