Simulator cache not persistent across runs

[yngve-sk]

What happened? (You can include a screenshot if it helps explain)

Since we now get simulator results from ERT storage, we should (if the experiment exists within the optimization output folder) be able to load an existing experiment, so that it can be searched for cached results.

Expected behavior:

1. Run a case normally from the CLI
2. Run it again and expect ROPT to skip all simulations as it already has the results

It is up for discussion how exactly we do this. Some things to consider:

• --new-run will wipe the storage clean, but we could add another option to keep the storage, or opt for keeping the ERT storage by default.
• This implies we only have one experiment per everest optimization case. If we have multiple experiments in the same storage, we need some mapping, which we currently have no setup for handling, so would opt for going with just one for now.

What did you expect to happen?

No response

steps to reproduce

Environment where bug has been observed

• python 3.11
• python 3.12
• macosx
• rhel7
• rhel8
• local queue
• lsf queue
• slurm queue
• openPBS queue

[yngve-sk]

Duplicate of #10037