change(scan): Create a scanner storage database, but don't use it yet

[teor2345]

Motivation

This does the database creation and column family parts of #8019, but doesn't actually use the database.

There's still the actual data part of that ticket left for the next PR, and testing.

PR Author Checklist

Check before marking the PR as ready for review:

• Will the PR name make sense to users?
• Does the PR have a priority label?
• Have you added or updated tests?
• Is the documentation up to date?

If a checkbox isn't relevant to the PR, mark it as done.

Complex Code or Requirements

This seems to just work as expected.

Solution

• Create an on-disk database in the scanner storage (but don't use it yet)
• Create a column family
• Use ephemeral storage in tests

Related changes:

• Move populating storage into the new() method
• Add a zebra_scan::init() method

Related fixes:

• Fix a bug in the state where format tests weren't being skipped on shutdown even if the database was configured to skip them

Testing

The existing tests found one bug already, we'll add more in the next few PRs.

Review

This seemed like enough code to review for now. It's not urgent but it is on the critical path to completing the scanner MVP.

Reviewer Checklist

Check before approving the PR:

• Does the PR scope match the ticket?
• Are there enough tests to make sure it works? Do the tests cover the PR motivation?
• Are all the PR blockers dealt with?
  PR blockers can be dealt with in new tickets or PRs.

And check the PR Author checklist is complete.

Follow Up Work

The rest of #8019:

• Define key and result serialization
• Put the keys in the database
• Put the results in the database
• Document the database format

[oxarbitrage]

Looking great but i think the failing test need further investigation.

thread 'tests::scanning_fake_blocks_store_key_and_results' panicked at zebra-state/src/service/finalized_state/zebra_db/block.rs:65:66:
called `Option::unwrap()` on a `None` value
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
test tests::scanning_fake_blocks_store_key_and_results ... FAILED

https://github.com/ZcashFoundation/zebra/blob/main/zebra-state/src/service/finalized_state/zebra_db/block.rs#L65

It is clear that the test do not have a state but i don't see why and where tip() needs to be called.

We can just ignore the test to merge and deal with it later but i don't think that will be a good idea.

[teor2345]

It is clear that the test do not have a state but i don't see why and where tip() needs to be called.

It's being called from a format change, which should have been disabled in PR #8002, but I missed it:

thread 'tests::scanning_fake_blocks_store_key_and_results' panicked at zebra-state/src/service/finalized_state/zebra_db/block.rs:65:66:
called `Option::unwrap()` on a `None` value
stack backtrace:
...
   2: core::panicking::panic
             at /rustc/79e9716c980570bfd1f666e3b16ac583f0168962/library/core/src/panicking.rs:127:5
   3: zebra_state::service::finalized_state::zebra_db::ZebraDb::spawn_format_change
   4: zebra_scan::storage::Storage::new
   5: zebra_scan::tests::scanning_fake_blocks_store_key_and_results
...

That's fixed by commits baaa738 and d57efb7

[oxarbitrage]

Looks great, thanks.

The base branch was changed.