I do understand what your problems are and I also know what you can do about it. The problem is, that most of the solutions mean additional work for you.
Example scraping: There is only basic support for scraping in MC. MC scrapes from a handful of websites, with almost no options. If I want MC to scrap movie descriptions in another language - does not work. If I want MC to scrap audiobook informations - does not work. I have to go the forum and post a feature request, which will not be implemented because it is not important enough. I understand that. You can't code hundreds of different scrapers for different websites and categories, which may be outdated in 5 years. However, the users just need them. Manually scraping data for a huge collection is just not doable.
The solution is on the hand --> plugins. If I look at competition software like Kodi, I see that they are living of plugins/addons/skins/etc.
This is the solutions for many of your problems. You may have to adapt some of your architecture and it will take 2 or 3 years, but when it works the community will help you improving your application and boosting sales. I would suggest to provide an API for simple programming languages like Python.
Example file tagging: You often do unnecessary additional work. File tagging is an example of this. You support file-tags for different for formats like mp3 and co. However all of this is unnecessary work. The xml-sidecar files work of every kind of file-type. Support for specific file-types is not necessary. These are the solutions you need everywhere.
Feature Creep: There are many parts in MC that need to be updated. The 3d-view looks horrible, the visualization looks horrible, most of the styles look like the are stuck in time, etc. When I look into my library, all the covers look blury, even on my 1080p screen. I do not want to know, what it will look on a 4K screen. This can be fixed by one developer in a week. However, it is not done, even it has been an problem that persisted for the last 10 years and bothered all kinds of users, regardless if they are music listeners or movie watchers. Still it won't be fixed anytime. Because new features are more important. Instead we have now a feature called "Listening Test" which most users will use once and then never, because then they know how good their listening accuracy really is. And we also have a feature to scrap subtitles. With the only backdraw that it is not checksum based and therefore makes zero sense, because all the different file versions need different synchronizations and nobody wants to adjust these timing when watching a movie.
You should definitely fix the things that are already in MC. Not adding completely new features, but instead improving the ones that are already part of it. If one is added, it should have a fundament. Someone should have thought through the whole thing. Not like the listening test, or the subtitles scraping.
Real improvements: Since I started to use MC19, nothing really changed. The biggest change was the Modern Card Skin, which I already used before. Small features are sold as big features, like the "View Extras"-feature. Sorry, but this is only another database-entry with the paths for the extras for every file, which is accessible through the GUI. It is something that is quite simple from a technical perspective and does not require much coding effort. It is one of the 5 features that got a pinned thread. There are no real improvements any more, even if there are ton of things that would require a feature.
MC is to static: I have a lot of audiobooks in my library and they are categorized by genre and subgenre. However not all genres have subgenres. MC requires everything to have the same depth. So I either get rid of the subgenre, or I apply it to audiobooks that don't even have a subgenre. It should be possible for MC to skip depth levels if all elements in it are of the same type.
I also can't use arbitrary cover arts for genres. Cover arts are only uesable for files, artists, albums, series and seasons. Why can't I use them for other tags? That is a constraint that is not necessary.
Another unnecessary constraint are the limited input types for certain tags. I want to edit the Sub media types. There are no technical reasons why I should not be able to add a type.
I see these kind of constraints everywhere in MC and they are really annoying.
You don't listen: For instance scraping data in languages other than english. I haven't read a feature request more often than this one. It is in every feature request thread, it is even in this thread. Still it is not implemented for years. Technically we are speaking of a combo box in the "Get Movie & TV Info" dialog, where the user can select one language, and maybe a second one with a fallback language. You just have to change one parameter in the request. If the resulting description is empty or very short, use the fallback language. It is nothing that can't be implemented by one developer in one week. Even though it is simple and often requested, it is not done.
I think you are strongly biased in terms on what kind of feature request you implement. Audio and tv seem to be first class citizens. GUI, scraping, series and movies do get some alibi improvements where you do small unimportant things, so at the end you can say you did something. Images and plugin support are locked into the cellar.
I think you should reprioritize this list. Audio already has really good support. There is no need to implement more features in this area. TV usage is declining. Video flatrates like Netflix and Amazon Prime Video are increasing. Why do you want to ride a dead horse?
On the other hand the plugin support, a good looking gui and many scrapers are the things competing software like Kodi have. This is where you should invest your time. In things where you are years behind the competition.
