Refactor: Where possible, utilise native xq flags for XML parsing and extraction

[Crosenhain]

This PR updates the XML parsing logic across all scripts (Lidarr, Radarr, Readarr, Sonarr, and Universal) to use xq's built-in extraction flags rather than piping outputs through jq, wherever possible. Where it isn't possible (eg. where jq handles things better), tell xq to output json using the -j flag.

Key Changes:

• Replaced some jq pipelines with xq -x for direct XPath queries (e.g., extracting Plex versions and config.xml values).
• Use the -j flag for xq to output JSON, so that jq can be utilised to parse Plex library data.
• Simplifies the command structures by reducing reliance on secondary jq processes where xq can handle the extraction natively.

[Crosenhain]

Hmm this actually appears to be a difference in xq versions between my lidarr and *arr instances that caused me to look at this in the first place...

[Crosenhain]

I dug a bit more, the xq and yq packages are probably the wrong ones to use.

I did discover that the yq-python package allows the original functions.bash script to function correctly in all cases, so that may well be the solution.