Add a model for managing permissions (power levels) in rooms. This gets rid of a whole bunch of boiler plate in NeoChat and as a bonus makes it easy to add a feature to allow setting the permission level for any event.
A category won't be shown if there are no relevant IDs (will add the ability to add new ones later).
Part of network/neochat#565
This draws heavily on what @carlschwan did in network/neochat!1579 but I found it easier to start again and grab the bits as I needed them plus some other copying from what I did in the Space tree model.
From my current limited testing this seems to work nicely try and break it.
At its best, this worked ok-ish, though it was always missing basic features.
It's also a massive memory leak and significantly complicates the codebase.
(Which is not yet cleaned up by this commit)
Currently, it is entirely broken and noone noticed or cared enough to report or fix that.
BUG: 455984
This adds previews for downloaded pdfs and code files.
Move the itinerary model representation to it's own component and instantiate from MessageComponentModel. This starts to lay some groundwork for previewing other files.
These share the same D-Bus service name (org.kde.neochat) which comes
with a fun little addition: KRunner activation! While this is not a
problem while NeoChat is running - since it's already registered - this
becomes an issue while searching for NeoChat in something like the
Kickoff. The Kickoff (and consequently, KRunner) tries to activate the
NeoChat D-Bus service which runs our unified push parts.
This introduces a "FakeRunner" which watches closely for calls to the
KRunner interface while we're in unified push mode (or directly called
from D-Bus but not running) so it quits immediately.
The aim is to be able to use separate delegate for things like codeblocks and quotes so that they can be styled differently.
Currently, this is only to pause playback of other media when something new starts playing.
In the future, it will also be used for other things, like call ringing, etc.
This is the first step in separation, so we can focus more on the
actions in this menu, and it's not tangled up in how the context menu is
shown and displayed.
For now everything should look identical. However this moves to using a model for the content of the message and is intended to lay the foundation for improved message content representation, e.g. splitting up a text message in multiple sections and using different delegates for things like code and quotes.
