[Metadata] Deal with anti-scraper measures

[nephros]

SailfishOS VERSION: N/A

HARDWARE: N/A

SailfishOS:Chum GUI application VERSION: <= 0.6.11

BUG DESCRIPTION

In the age of AI Companies, many git hosters and other sites have adopted tools like Anubis to thwart their irresponsible scraping.

These scrapers break Chum GUI app desciptions where URLs are used, e.g. the DescriptionMD tag, and sometimes screenshot and icon links.

STEPS TO REPRODUCE

As an example- look at the app page of the waypipe package.

1. Have a package which uses DescriptionMD in the Chum Metadata
2. Browse to the App description page in Chum GUI
3. Observe some raw html snippets instead of the markdown content
4. Alternatively, try curl -A "Mozilla/5.0 actually curl "-L https://gitlab.freedesktop.org/mstoeckl/waypipe/-/raw/v0.10.4/README.md to see Anubis output.

ADDITIONAL INFORMATION

[nephros]

... in the case of Anubis, the effect may be lessened by using a User-Agent
which does not contain the word Mozilla.
But strictness is up to the configuration of the respective site.

But setting the User Agent in a QML XHR is supported only in very recent versions of SFOS' Qt:

• Anubis default config: https://github.com/TecharoHQ/anubis/blob/main/data/botPolicies.yaml
• Remove user-agent filter sailfishos/qtdeclarative#9