* refactor and clean up diagnostic engine code so that it can be re-used stand alone.

code refactored is related to properly set up CompilationWithAnalyzer and compute and return diagnostics

* made default diagnostic analyzer service to share code with VS diagnostic analyzer service on computing diagnostics.

unlike VS diagnostic analyzer, the default one doesn't persist data or calculate diagnostics per a analyzer to improve per analyzer freshes nor compute and maintain per project analyzers and etc.

also, this add ability to calculate semantic diagnostics for script file.

* enable semantic diagnostics for script file for misc workspace.

now, C#/VB script files in misc workspace will get semantic errors as well as syntax errors.

any language such as F# if document is added to misc workspace with SourceCodeKind.Script, they will automatically get semantic errors as well.

this PR doesn't address changes propagations for #load which was never supported for existing C#/VB script support.

* addressed PR feedbacks

Standardize the conditional IOperation verification skip method, and fix a bug that caused it to skip in exactly the wrong way.

* Set working folder for the solution added.

persistent service requires working folder location to find out where to save its persistent file

* added integration test to verify working folder exist

It was holding onto a SolutionInfo directly and then throwing away
the non-SolutionAttributes portion of it, only because when the code
was originally written SolutionInfo.SolutionAttributes didn't exist.
It's silly now.

Merged on behalf of @maca88

* Added UseOriginalEndOfLine option on SyntaxRemoveOptions to preserve the EOL from the original node (#22924).

* Grammar correction.

* Removed the normalizer parameter

* Removed UseOriginalEndOfLine from shipped api

* Added UseOriginalEndOfLine to unshipped public api

* Added a null check in AsRootOfNewTreeWithOptionsFrom method

* Removed HasEndOfLine method in favor of GetEndOfLine to avoid duplicate calls.

* Modified KeepEndOfLine option to add an EOL for a directive only if there is one in the original document

* Modified KeepEndOfLine option to preserve the EOL from the original node and removed UseOriginalEndOfLine option.

* Modified AddEndOfLine method in order to avoid code duplication

* Minor modifications

* Added more tests and removed unneeded GetEndOfLine calls

* Added a static method to test both Windows  and Unix EOL and used where it matters.

* Added a fallback EOL when requiresNewLine is set to true

Just removing a TODO.

* Use Arcade testing targets

* Skip CodeStyle tests on Mono

* Include net472 in test target frameworks for Mono

* Remove incorrect file existence check.

* Update ADO integration test yaml

This change simplifies our data flow analysis by removing an obsolete subclass and renaming some of the components to standardized names used in the literature on lattice data flow analysis.

Fixes #23207 

* some code clean up and added some code instruments to make perf investigation easier

* changed lambda to local function and addressed feedback

* Classify malformed XML Doc comments as XmlDocCommentText

* Added test for malformed xml doc comment classification

* Updating classification for discard designation and underscore tokens to be Keywords rather than Identifiers.

* Removing underscore from the IsKeywordKind check, since it is covered by isContextualKeyword

* Reverting change to SyntaxFacts and moving discard check back to ClassificationHelpers.

* Reverting all changes to SyntaxFacts.

* Add a comment clarifying how _ is classified.

make sure to VS and OOP to use same file to get mvid of analyzer file reference when creating checksum for both VS and OOP (#31438)

the problem is that the OOP always used the actual file that got used for analyzer (in other words, shadow copied file in VS), but VS used original file to get MVID.

based on shadow copying logic, 2 (with same assembly identity) might actually be 2 different files by sharing a shadow copied file between 2 original files with same assembly identitis causing MVID to be different.

this make sure all use same file for MVID so that regardless of host implementation of analyzer loader (who is the actual thing that control what file will be used to get analyzers), we get consistent MVID

These methods are called any time we add a new project and we have a
new output path; this was causing us to realize a Project object for
every project in the solution, which was O(n^2) during solution load.
This was adding a few hundred milliseconds of overhead in the immutable
data structure, which in this particular case we don't need.

Fixes dotnet/roslyn#31306.

This changes RunTests to always use procdump when tests timeout.

Previously there was a single option, procdumppath, which controlled
both monitoring xunit processes with procdump and using procdump on time
out. The former causes stability issues at times with test execution and
has to be disabled in a couple of places (spanish and integration
tests). The latter has no stability issues and hence should always be
used.

Related issue: https://github.com/dotnet/roslyn/issues/25145
This PR add the function to VS and let use to pull members up via quick action

Address feedback from https://github.com/dotnet/roslyn/commit/dacca3d1bd6509935e4d01cf3f0e242075c217f2#r31467643

// "x is int _" => "x is int"

Address feedback from https://github.com/dotnet/roslyn/commit/dacca3d1bd6509935e4d01cf3f0e242075c217f2#r31467539

// "M(out var _)" => "M(out _)"
// "M(out int _)" => "M(out _)"

Add unit test for https://github.com/dotnet/roslyn/commit/dacca3d1bd6509935e4d01cf3f0e242075c217f2#r31465616 - unable to repro it locally

Fix code fixer for replacement inside a switch section. Addresses feedback from https://github.com/dotnet/roslyn/pull/31387/files#r236553196.

[CHANGED] Removed the filepath from the filename of the log. The loggingFileName could be "C:\Users\user\build" which means that the second string in Path.Combine generates to "server.C:\Users\user\build.log" and that will be combined with the path once again leading to a full file path of: "C:\Users\user\build\server.C:\Users\user\build.log" and I dont think thats legal on Windows.