MKDocs conversion (#278)

Closes #252, Closes #250

## Changes
- ReplayWeb.page uses MKDocs for documentation now!
- Requires both
[`mkdocs-macros`](https://mkdocs-macros-plugin.readthedocs.io/en/latest/)
and [`mkdocs-redirects`]
(https://github.com/mkdocs/mkdocs-redirects) plugins to be installed
- Redirects existing docs pages to their new locations under either
`/embedding` or `/user-guide`
- Unused images have been removed
- Adds new ReplayWeb.page branding!

All pages have received copy edits. All references to future (or
previously) planned features have been removed. More major content
changes of note are as follows:

### Supported Locations
- Made into a page, no longer a table. Locations are either supported or
not listed.

### Offline Use 
- Removes browser cache section, adds improved PWA install and uninstall
instructions.
- Removes navigation instructions to GitHub's releases section in favour
of a link

### Supported Formats
- Moved to the User Guide homepage
- Updated the format support

### Exploring Archives
- Example links for showcasing features have been moved to their
respective feature documentation section on this page

### Remove Usage, Examples, and Troubleshooting Pages
- Usage content is covered in the user guide index
- Troubleshooting content has been added to a question section on the
getting started page about WARC vs WACZ.
- Examples are now Featured Projects on the docs homepage!

### Build Process
- Removes Jekyll & related site assets.
- Adds new GitHub action for publishing
- Adds a python macro to dynamically load the replayweb.page version from package.json
- Relocates replayweb.page site content under /site (including docs)
- Some assets (favicon, logo) and sw.js/ui.js automatically copied into /site

---------
Co-authored-by: Ilya Kreymer <ikreymer@gmail.com>

site/docs/user-guide/locations.md

@@ -0,0 +1,27 @@
+# Supported Location URLs
+
+ReplayWeb.page uses URLs to uniquely identify web archives.
+
+In addition HTTP/S, the system supports loading from additional cloud providers and via peer-to-peer decentralized services. The goal is to support loading web archives from any source a web browser can connect to.
+
+The following is a list of currently supported URL schemes and details below:
+
+## HTTP and HTTPS
+
+While `https://replayweb.page` can load archived items from any URL, the hosting server must allow access by [enabling cross origin resource sharing](../embedding/cors-settings.md) (CORS). HTTP URLs may not be accessible from `https://replayweb.page` since it requires HTTPS.
+
+## Local Files
+
+Local files selected via the UI file chooser, and will be given a `file://` URL. Unlike other web archive URLs, file URLs can not be shared with others.
+
+## Amazon S3
+
+The `s3://` URL scheme is provided as a convenience to reference S3 URLs. The URL is converted to an HTTPS URL for access. The S3 Bucket must be public and also provide access to `https://replayweb.page` via CORS. Downloading via S3 credentials is not currently supported.
+
+## Google Drive 
+
+The `googledrive://` URL scheme is used when accessing a web archive shared over Google Drive. It provides a unique identifier provided via Google Drive. When accessing a Google Drive URL, users will be asked to authenticate to be able to access the file.
+
+## IPFS
+
+The `ipfs://` URL scheme is provided as a convenience and resolves to Cloudflare's IPFS Gateway to download archives from IPFS.