Add support for validating functions inside schema's rules

[Calyhre]

Is this adding or improving a feature or fixing a bug?

Adding a feature

What's the new behaviour?

With this change we can now use validating functions inside schema's matching properties, at least for a few keys.

A few example:

{
  match: { /* something */ },
  object: kind => kind !== 'text',
  type: type => ['paragraph', 'image'].includes(type),
  data: data => data.has('thing'),
  marks: mark => mark.type === 'mention' && mark.data.has('userId'),
}

I also added documentation, specs, and default normalization on next and previous rules.

How does this change work?

A more complex one now allowed by this change:

{
  // Merge two consecutives lists
  match: { type: 'list' },
  next: {
    type: type => type !== 'list'
  },
  normalize: (change, error) => {
    if (error.code === 'next_sibling_invalid_type') {
      change.mergeNodeByKey(error.next);
    }
  },
}

Have you checked that...?

• The new code matches the existing patterns and styles.
• The tests pass with yarn test.
• The linter passes with yarn lint. (Fix errors with yarn prettier.)
• The relevant examples still work. (Run examples with yarn watch.)

Does this fix any issues or need any specific reviewers?

Fixes: #2139

[ianstormtaylor]

Thank you @Calyhre, this is an absolutely perfect pull request!