Code Refactor - Pass 1

[BryceBarbara]

This is one of the stepping stones on the road to modernizing the codebase. Beyond this, I think the last thing to do before the real work can begin is enable nullable reference types and fix up any issues from it.

Changes

• Converted the project to a .NET 5 project
• Target "Any CPU" rather than "x86" specifically
• Split up AnmDecoder into files specific to each supported engine
• Converted a number of files to use Spans rather than byte arrays, since Spans avoid having to pass around start offset/length everywhere
• Fixed some of the issues from the previous merge
  • Didn't use LevelObjectPropertiesArea control
  • Duplicated StringExtensions.cs file

[bigianb]

There's quite a bit of repeated code now- I assume that will be refactored out later and will be made more DRY. I think the span implementation is quite in-efficient but as it's not used in any performance sensitive code that's fine.

[BryceBarbara]

Yeah, repeated code isn't ideal but there are so many lower hanging fruit that I want to address prior to that. The primary goal of this PR was just to get .NET 5 working.

And I agree about the span stuff. Calling ToArray is a terrible practice and it pretty much undoes any performance benefits we might have had. But I still see it as a benefit because it simplifies code since it doesn't start offsets. Later, we can replace Span with some other type that doesn't have performance hits but has the same behavior of making a subset of an array and passing it around without actually duplicating memory.