Run benchmarks on mock chain

[abailly-iohk]

Everything is in the title: This PR resurrects the mock-chain to be able to run larger Hydra clusters.
I plan to address the actual distribution of the Hydra cluster over several machines in a latter PR, possibly with the support of Ryan Williams, should we think this is the way to go.

[KtorZ]

👍 we would want the benchmark to support running on either the mock-chain or the real chain, even though it doesn't change much for the benchmark results I think we still want to drop ZMQ in the long-run once we have a better benchmarks infrastructure.

[abailly-iohk]

In view of recent developments, I wonder if this PR still makes sense:

1. It adds the complexity of more parameters which would be only relevant for benchmarks purpose, which is annoying,
2. it seems deploying an actual cardano private network might be a more interesting experience anyhow. In my mind it would have made more sense in relationship with ADR-0015 - Externalise chain server #230 but 🤷,
3. It's not super urgent to have that 50 nodes benchmark, we can wait for the infra and cardano nodes setup Ryan williams is working on,
4. It would be more fruitful in the long run to have the benchmarks be parameterised by run duration instead of number of transactions, which means making them more "dynamic": Generate transactions and store results online instead of in a single step at beginning and end of the benchmark,
5. I don't want to fall prey to "Sunk Cost Fallacy" and spend even more time fixing it.

I am going to close this, we can always reopen it if @ch1bo or @KtorZ think it's still worthwhile.