Due to scenarios wherein a formatted link ended up as part of a larger
raw link after parsing, change the containment check to an overlap check
and add appropriate tests for these edge cases.
Testing with #6542 surfaced a crash scenario, caused by formatted links
that had URLs in the display text, for example
[mean example - https://osu.ppy.sh](https://osu.ppy.sh)
In that case the outer Markdown link would get picked up once, and then
reduced to the link text when looking for other links, leading to it
being picked up again the second time when the raw link is found.
Add a check in the raw link parsing path that ensures that the found
URL is not a part of a bigger, pre-existing link.
While reviewing #6542 it became apparent that there was another Markdown
link format variant, used in comments that came from the web API, called
the "inline link" style. It allows to specify the tooltip title within
the actual URL portion, as such:
[link text](https://osu.ppy.sh "tooltip text")
Add tests with a couple of easy and trickier examples of such a format.
Moreover, add a new edge case of a Markdown link with a link inside
the display text, which during tests was detected to be problematic.
Add a test case reproducing the conditions encountered "in the wild"
wherein a skin import would be performed incorrectly due to a __MACOSX
resource fork directory present next to a directory with the actual skin
files in the archive.
As a purely cosmetic code improvement, substitute string literals
in constructor calls of HeadlessGameHost in ImportBeatmapTest for nameof
operator usages.
Tests for the line-buffered reader added in 7b1ff38 were subtly
dependent on the execution environment due to differing end-of-line
markers on Windows and Unix-based systems.
Because StreamReader discards all newlines when reading line-by-line,
LineBufferedReader used a StringBuilder to patch the peeked lines
back together with the remaining contents of the file being read.
As StringBuilder.AppendLine uses the environment-specific newline
delimiter, the delimiters after the peeked-but-unconsumed lines can
therefore be substituted by the platform-specific variants, causing
the test failures due to the overly-simplified way they were written.
Reformulate the test to avoid such issues from resurfacing again
by splitting lines by \r or \n and then testing each line individually.
Additionally remove all raw literals in favour of explicitly mixing
various line delimiter character sequences for additional coverage.
