create-testnet-data: correctly compute set of credentials delegating votes to a drep in conway genesis

[smelc]

Changelog

- description: |
    create-testnet-data: correctly compute set of credentials delegating votes to a drep in conway genesis
# uncomment types applicable to the change:
  type:
  # - feature        # introduces a new feature
  # - breaking       # the API has changed in a breaking way
  # - compatible     # the API has changed but is non-breaking
  # - optimisation   # measurable performance improvements
  # - refactoring    # QoL changes
  - bugfix         # fixes a defect
  # - test           # fixes/modifies tests
  # - maintenance    # not directly related to the code
  # - release        # related to a new release preparation
  # - documentation  # change in code docs, haddocks...

Context

There was uncertainty (see this thread on Slack and here on GitHub) when fixing some part of create-testnet-data when writing #940. Because we wanted to merge 940 fast, we proceeded leaving an incorrect value in create-testnet-data's conway genesis, in the presence of dreps; because we needed to synchronize with the ledger team to understand what to do.

This PR corrects the incorrect value.

How to trust this PR

Here is a before and after comparison of the conway genesis created by create-testnet-data:

On the left, one can see that the delegators fields of the initialDReps are all left empty. But this is incorrect: the field delegs (above in the screenshot) show that indeed there are votes delegations happening.

On the right, you can see that the delegators fields are populated with values corresponding to the ones in delegs ✔️

Checklist

• Commit sequence broadly makes sense and commits have useful messages
• Self-reviewed the diff

[smelc]

Note to @palas: this new delegs value is like the old cgDelegs one, except that it doesn't hardcode applying some constructors right away in nested positions. This allows to use delegs at the source for computing cgDelegs and cgInitialDReps. This is important to make sure we are consistent between the two.

[Jimbo4350]

Great work. The screenshot was perfection 👌.

[Jimbo4350]

This is problematic and thankfully it showed up in the tests:

/out/conway-genesis.json
        149 ┃ 
        150 ┃     length (cgInitialDReps conwayGenesis) H.=== numDReps
            ┃     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
            ┃     │ ━━━ Failed (- lhs) (+ rhs) ━━━
            ┃     │ - 4
            ┃     │ + 5
        151 ┃ 
        152 ┃     length (cgDelegs conwayGenesis) H.=== numStakeDelegs

We have: numStakeDelegs = 4 and

numDReps :: Int
numDReps = 5

Because we use zip the 5th drep is ignored because there is no stake key to delegate to it. We must therefore take into consideration dreps with no delegators.