Make identity store loading and alias merging deterministic

[banks]

Description

To optimize loading entities into the IdentityStore during unseal we current load the 256 buckets from storage in parallel. We then process them in whatever order they load. This determinism should be fine because each entity should be loaded by ID and so the loading order shouldn't matter.

But in real-life historical (and potentially current) bugs can cause there to be duplicated aliases in storage. We have for many years attempted to cleanup and merge such problematic duplicate aliases on load, however we always merge them in the order they are encountered during loading. Because this is non-deterministic, it means that different aliases can "win" this merge process after different unseals causing unpredictable behaviour. It will often also mean that Enterprise Performance Standbys may end up with a different view of the entities than the active node causing inconsistent results depending on which node responds to a request.

This PR retains the parallel loading optimization but fixes the order of processing loaded buckets to ensure that all nodes will resolve any duplicates identically.

We've reviewed this in the Enterprise PR extensively and performed performance testing that shows that even though there is a theoretical worst-case that might make this new approach slower (say if the first bucket takes a long time to load), in practice it's not measurably different (mainly because the current code doesn't realise ideal parallelism anyway due to contention in other layers of storage).

This should fix issues where duplicates (caused by other bugs) then cause inconsistent responses from different servers.

JIRA: VAULT-31384
Ent PR: https://github.com/hashicorp/vault-enterprise/pull/6776
RFC: https://docs.google.com/document/d/16Tbsngmzg9tuJu1G8s1uSJGDvp-YS5UUUVVboDvuQpc/edit?tab=t.0

TODO only if you're a HashiCorp employee

• Backport Labels: If this PR is in the ENT repo and needs to be backported, backport
  to N, N-1, and N-2, using the backport/ent/x.x.x+ent labels. If this PR is in the CE repo, you should only backport to N, using the backport/x.x.x label, not the enterprise labels.
  • If this fixes a critical security vulnerability or severity 1 bug, it will also need to be backported to the current LTS versions of Vault. To ensure this, use all available enterprise labels.
• ENT Breakage: If this PR either 1) removes a public function OR 2) changes the signature
  of a public function, even if that change is in a CE file, double check that
  applying the patch for this PR to the ENT repo and running tests doesn't
  break any tests. Sometimes ENT only tests rely on public functions in CE
  files.
• Jira: If this change has an associated Jira, it's referenced either
  in the PR description, commit message, or branch name.
• RFC: If this change has an associated RFC, please link it in the description.
• ENT PR: If this change has an associated ENT PR, please link it in the
  description. Also, make sure the changelog is in this PR, not in your ENT PR.

[github-actions]

CI Results:
All Go tests succeeded! ✅

[github-actions]

Build Results:
All builds succeeded! ✅

[mpalmi]

Looks great, Paul! Just FYI: I cherry-picked the squashed commit in enterprise and it builds just fine.