Expose --force-fallbacks to the CLI

[ThomasDalla]

The rocketpool binary itself supports forcing the use of the fallbacks.

--force-fallbacks
Set this to true if you know the primary EC or CC is offline and want to bypass its health checks, and just use the fallback EC and CC instead

It would be useful to expose that flag to the CLI so that we can run commands using the fallbacks.

For example, currently the latest version of Besu has known compatibility issues with the latest version of the smartnode.
Notably when we try to claim rewards, we get an RPC error:

$  rocketpool node rewards
Could not get node rewards: Requested range exceeds maximum RPC range limit

If we have a fallback using a different EL, we could simply run rocketpool node rewards --force-fallbacks so that it works, but because the flag is not exposed, we cannot.
A workaround is to temporarily switch the primary and fallback endpoints in the config, but it would be more convenient to pass the --force-fallbacks flag when running the command.

Alternatively, on RPC errors, rocketpool could automatically use the fallback node, which could be an even smoother end user experience.