Deprecates implicit cache keys

[arcanis]

What's the problem this PR addresses?

The lockfile has a very small optimization feature: instead of repeating the cache key across all checksums, it keeps a global cache key at the top of the file, which it uses everytime a checksum is missing a cache key. This saves a few bytes.

However, I found out this behaviour causes breakages under some situations:

• Alice creates a feature branch from main and add a new dependency foo
• Bob upgrades Yarn to a new version that compiles zip files differently
• Alice merges the new main into her feature branch
• Git merges the two lockfiles together; let's say there are no conflicts
• The foo entry added by Alice has no cache key, since it was added at a time its cache key matched the global one
• However it doesn't match the new global one, which was updated at the top of the file

How did you fix it?

The checksum cache key is now always set.

Checklist

• I have read the Contributing Guide.

• I have set the packages that need to be released for my changes to be effective.

• I will check that all automated PR checks pass before the PR gets reviewed.