Allow Gateways to Specify the Selection retry Timeout

[rickstaa]

Is your feature request related to a problem? Please describe.

After #3033 is merged the Gateways tries for 2 seconds to find compatible Os that can do a AI job. Currently this value is hardcoded. We might want to give Gateways a way to set this value in the future.

Describe the solution you'd like

I would like to add a CLI flag that Gateways can use to increase or decrease the default value.

Describe alternatives you've considered

I did not yet consider alternatives.

Use Case

No response

Expected Outcome

No response

Additional context

No response

[rickstaa]

Closed in favor of a better communication protocol where the gateway communicated with the dApp about how far a job is. Dapps can then decide themselves to disconnect and try again.