Motivation

Refactors and improvements of the markdown parsing system:

• Extract markdown-related code + tests in dedicated file
• Clearer API surface
• Title found in markdown content (not frontmatter) is returned as a separate attribute, not added to frontMatter.title => ability to handle those cases separately
• Ability to extract title from markdown documents using the alternate markdown syntax (title\n===)
• Fix empty excerpt when Markdown doc has a title: excerpt should simply skip/ignore the title
• Duplicate title warning: better wording + ability to disable it (as it can be annoying for some sites after upgrading, cf RN website)
• Initial code to have more strict frontmatter validation using Joi (to be completed)

Have you read the Contributing Guidelines on pull requests?

yes

Test Plan

tests

Related PRs

Follow-up of #4485 cc @armano2

Also related to #4582 (as we want to support better sites without any frontmatter at all)