Improve debug forking

[evaporei]

I do accept any suggestions on the CLI API, I'm not sure those parameters should be passed where I've put them (and with those names).

This PR adds:

• --debug-fork CLI option
• --start-block CLI option

With those you can basically get any subgraph from the Hosted Service and debug it at any block without having the source code to change the start block, which debug forking currently requires (docs).

Eg (enzyme fork that I've been able to reproduce #3236) :

cargo run -p graph-node -- \
          --postgres-url $THEGRAPH_STORE_POSTGRES_DIESEL_URL \
          --ethereum-rpc $ETHEREUM_RPC_URL \
          --subgraph QmTBxvMF6YnbT1eYeRx9XQpH4WvxTV53vdptCCZFiZSprg \
          --ipfs $IPFS_URL \
          --debug-fork QmTBxvMF6YnbT1eYeRx9XQpH4WvxTV53vdptCCZFiZSprg \
          --start-block 0x8e28f0485a3bbcca851d502c2ca1120607b7178bf997ca8a40f4a457b8acb964:14178262 \
          --fork-base $FORK_BASE_URL

I know a lot of parameters, that's why I'm not sure if this is the way to go with the API design.

@VIVelev if you can give any suggestions here, thanks a lot your tool is super useful ❤️