Standardize and Optimize Code with dotnet format

[TiKevin83]

Follow up on #3681

Proposes merging the corrected output of dotnet format style -v detailed --severity info, optimizing and/or simplifying code per dotnet standards and removing unused code.

Implements or standardizes the following:

• destructuring assignment including in foreach initialization and in combination with tuples for concurrent assignment operations, removing the need for temp variables
• shorthand new() that doesn't repeat the variable type
• new switch syntax
• new using syntax
• new local function syntax
• |=, &=, >>=, <<=, ++, --, +=, -=, %=, ??= operator usage
• consistent readonly usage
• removes unused/commented out code
• lots of simplified boolean logic
• use of the is and is not keywords in favor of == and != where relevant
• array access from end with [^]
• array range operator with [..] (also simplifies many substring uses)
• optional chaining with ?.

potential outlying changes:
There's a change to RCheevos.cs that looks possibly intended to be the existing way
Some removals of unused code in the Amstrad CPC, Commodore 64, PC Engine, and Sinclair Spectrum CPU cores may not make sense (it looks like a bunch of cpu registers end up unused)

Check if completed:

• I have run any relevant test suites
• I, the committer, have read the licensing terms for contributors (last updated 2022-07-15) and am compliant

[YoshiRulz]

This would be a good idea to enable—after the codebase has been given a manual once-over. This way is like seeing a dead canary and deciding to pump clean air into the birdcage.

[TiKevin83]

This would be a good idea to enable—after the codebase has been given a manual once-over. This way is like seeing a dead canary and deciding to pump clean air into the birdcage.

I don't think the order is that crucial. I did look briefly before and found I can definitely improve the codebase with a manual review in ways that the auto checks can't find with things like LINQ expressions, so I could work on that first though if preferred. This PR also doesn't strictly enforce following dotnet format as a pre-commit or CI check which could be desired if the manual combing is also in place.

[tommai78101]

Would it be helpful to add some info about the formatting style to the CONTRIBUTING file, and also the suggested / recommended way of checking for the formatting style based on the outputs of the dotnet style? https://github.com/TASEmulators/BizHawk/blob/master/contributing.md

[TiKevin83]

Would it be helpful to add some info about the formatting style to the CONTRIBUTING file, and also the suggested / recommended way of checking for the formatting style based on the outputs of the dotnet style? https://github.com/TASEmulators/BizHawk/blob/master/contributing.md

Definitely would be helpful to add that to Contributing however "format" is a bit misleading here, this PR is less intended to change formatting standards like indent style and more making use of new language features where the format command happens to know how to implement them automatically.

[tommai78101]

I see. I look forward to seeing the new info added to CONTRIBUTING.

[Morilli]

These should be reverted, as even though it's technically a "simplification" it makes the code less readable. The resharper disable is already evident of this.

[TiKevin83]

I disagree on this one, usually something is less readable/over complicated if you're returning true in the true condition of a ternary or likewise false in the false condition, but if someone is already intentionally ignoring the optimization then sure we can revert it

[TiKevin83]

Rebased on Morilli's PR and reran against the analyzer setting changes. I think something might be wrong with the IDE0007/IDE0008 var vs explicit settings still, will play with that as I have time.

[Morilli]

I think the main issue here is that a lot of the inspections at suggestion level only really work as just that: A suggestion.

Blanket-applying those everywhere unconditionally may not be desired in all cases.

[TiKevin83]

for sure I'll toy with it more, I'll probably rerun it at warning level and retune individual rule warning levels if I see that missing more than I think people would agree on