feat(docs): Add Chinese Localization Update for Tact

[sansx]

Issue

This PR resolves the Chinese localization tasks for Tact, which related to issue #934. The entire translation has been completed using Crowdin's AI-powered translation tool.

Preview Link:
https://tact-git-l10nmain-townsquarexyz.vercel.app/zh-cn/?_vercel_share=iQ0dtxO3eY4jES4Nl0IuLiK7ll6apcut

Notes:
The translation is fully AI-generated and will require proofreading to ensure accuracy and contextual appropriateness.

We'll provide further proofreading in future updates. Additionally, if anyone is interested, you are welcome to join our Crowdin project to help with proofreading and finalizing the translations.

Checklist

• I have updated CHANGELOG.md
• I have run the linter, formatter and spellchecker
• I did not do unrelated and/or undiscussed refactorings

[novusnota]

Great, thanks for the PR! Regarding formatting issues in crowdin.yml, run the yarn fmt:check -w to resolve those.

Because of the PR from the org, Tact team cannot really edit your branch, so I'm going to provide suggestions in text form only:

1. Double-check that all links lead to proper places — running yarn build inside docs/ folder will build the docs AND check that all the internal links point to places that exist. This is especially important since AI can hallucinate them on the fly — that's probably the reason why there was a new anchor link, that failed the CI.

2. Double-check that formatters on all pages match the following schema: https://starlight.astro.build/reference/frontmatter/. Locally running yarn build would find any issues.

3. Double-check that MDX components, such as Steps, Tabs, Asides (with ::: syntax), Link Cards and the like continue to work the same. Locally running yarn build would find any issues.

Points 2 and 3 seem to be fine from my viewpoint, but double-checking is always a plus :)

Mind you, that there are LOTS of pages in your PR where English and Chinese sentences or paragraphs are mixed together or follow each other. I'm not sure that is a good thing, because if the whole page remained in English, one could've always just run Google Translate or other website translation tool and get decent results. But with mixed content, that's no longer an option — so it's probably better to exclude such pages until a proper translation is present there.

You don't have to provide translations for all pages - the documentation automatically generates fallback English pages with a note at the top about the missing translation. And to add to my point about website translation tools, I believe that when it comes to good localization, quality is much more important than quantity. So if you are able to provide only 10 complete and good translations, it's much better than 100 lackluster AI ones :)