`restore-keys` is a configuration option for the actions/cache action
which specifies fallback behavior. The [docs][docs] say it best:
> When a cache miss occurs, the action searches for alternate keys
> called `restore-keys`.
>
> If you provide `restore-keys`, the `cache` action sequentially
> searches for any caches that match the list of `restore-keys`.
> ... If there are no exact matches, the action searches for partial
> matches of the restore keys. When the action finds a partial match,
> the most recent cache is restored to the `path` directory.
So this improves caching when there's a miss. For example if I edit
`.github/workflows/languages.toml`, the current behavior is that the
cache for downloaded grammars will miss and all of them will need to
be fetched again. With `restore-keys`, we use the latest published
cache as 'good enough', we'll fetch whatever grammars changed, and
then at the end we publish a new cache under the new hash.
[docs]: https://docs.github.com/en/actions/using-workflows/caching-dependencies-to-speed-up-workflows#example-using-the-cache-action