We will enable this rule when we migrate h1
titles to the title front matter attributes.

The SVG replacement codes can't be guaranteed to be
real words, but it doesn't matter. They shouldn't be
linted by Vale. This MR proposes a TokenIgnores
line that skips items matching a regex of **{this}**.

Note that tier badges, blessedly, use **(this)**
instead of curly braces, so we can lint them
separately.

The .vale.ini file for linting tests contained a section for
instructions on how to lint just *.md files, and another section for
linting both *.md and *.txt files. If you included conflicting
instructions in those sections, Vale became very confused and did
not apply tests in the manner you might expect.

This commit removes the redundant section so test results will be
predictable and less error-prone.

Vale (https://errata-ai.github.io/vale/) is a grammar, style, and word
usage linter for the English language. This merge request adds an
INI file for Vale (.vale.ini) to the root directory of the repository,
and creates a hidden folder (doc/.linting) for the storage of linting
tests.

Since the TW team uses multiple linting tools, and Vale also supports
multiple collections of rules (called 'styles'), the path to the tests
folder is doc/.linting/vale/styles/gitlab to support any future
expansions that can't be predicted at this time.

The folder currently contains only a .gitkeep file to meet the goal
of a MVC. Tests should be developed, vetted, and added in separate
merge requests, but those MRs all depend on the addition of the
INI file and the directory structure.