JDC should stop sending shares upstream if pool prev hash is different from JDC prev hash

[Fi3]

Right now when the pool send set new prev hash to the JDC, the JDC do nothing. The correct behavior should be:

1. compare it with the SNP it is mining on
2. if different stop sending shares upstream (in the future would be nice also to send a signal downstream to pause the miners)

    fn handle_set_new_prev_hash(
        &mut self,
        _: roles_logic_sv2::mining_sv2::SetNewPrevHash,
    ) -> Result<roles_logic_sv2::handlers::mining::SendTo<Downstream>, RolesLogicError> {
        warn!("SNPH received from upstream, proxy ignore it, and use the one declared by JOB DECLARATOR");
        Ok(SendTo::None(None))
    }