Problem

At the moment when the WORKER_MODE is set to teeracle the service spins up all teeracle services.
Shown here: Teeracle Service

This causes us to always launch, retrieve and submit oracle data for every oracle that we support.

Solution

I propose we add a new cli flag Teeracles which inputs a Vec<TeeracleTypes> into the RunConfig struct shown here: RunConfig in Service

This new argument can look like the following in practice:

./integritee-service run --teeracles teeracle1 teeracle2 teeracle3

Where each teeracle can simply be parsed as a string which can then be mapped to a TeeracleType

pub enum TeeracleType {
    Market,
    Weather
    ..
}

impl FromStr for TeeracleType {
    type Error = () // Some Error type maybe &'static str
    
    fn from_str(s: &str) -> Result<Self, Self::Error> {
        match s {
            "weather" => TeeracleType::Weather,
            // .. etc
        }
    }
}

// Then can do something with the clap::ArgMatches in
impl From<&ArgMatches<'_>> for RunConfig {
	fn from(m: &ArgMatches<'_>) -> Self {
		...
                 let teeracles = m.values_of("teeracles").map(|i| /* TeeracleType::from_str(i).unwrap_or_else(..) */).collect();
		Self { teeracles ..}
	}
}

This also requires editing the cli.yml and adding the teeracles field similar to how the interval is handled here: teeracle-interval cli.yml

subcommands:
    - run:
        about: Start the integritee-service
        args:
            - teeracles:
              required: false
              long: teeracles
              short: t
              help: List of Teeracles to launch Example --teeracles Weather Market SomeOtherOracle
              multiple: true
              takes_value: true
        # .. Other args here

We can then pass the list of teeracles to initiate to start_interval_update function in order to match on each TeeracleType and start the teeracle services accordingly to what we care for.