This seems to be a popular call among analyzers, especially XUnit. It's a
reasonable way to acquire a well-known analyzer type, but analyzers often
don't save the resulting symbol. This causes significant CPU work to be done
repeatedly as XUnit looks for the same symbol over and over, passing in the
same type name each time.

A very small, simple cache seems to solve the problem without requiring any
changes on the XUnit side.

Update SourceLink to 1.0.0-beta-63119-01

* make AbstractObjectBrowserLibraryManager constructor lazier

* PR comments

* readonly and use expression body

Might fix https://github.com/dotnet/roslyn/issues/28782

Fix "Run Code Analysis and Suppress Active Issues" command for CPS pr…

Put completion in 'suggestion' mode when typing a lambda to a method that could be an extension or instance method.

Do not qualify member accesses that appear inside attributes.

Disable leak detection tests on x64

Merge dev15.9.x to master

Looks like RyuJIT change causes our leak detection tests to fail.
Disabling until we find a better pattern to use there.

https://github.com/dotnet/roslyn/issues/28639

Use IHierarchyItemToProjectIdMap service to compare project hierarchies.
Fixes #27002

Replace Nonshipping property with IsShipping

Update Language Feature Status

Add tests.

Merge dev15.8.x to dev15.9.x

Merge dev15.9.x to master

Merge dev15.8.x to dev15.9.x

Fix a bad merge

These two PRs were green but broke when built together:

- https://github.com/dotnet/roslyn/pull/28652
- https://github.com/dotnet/roslyn/pull/28420

When we enabled pointer_element_access to work on fixed sized buffers
regardless of the movability of the receiver of the buffer access, we
did not repair the safety rules to prevent leaking a pointer to a
moveable variable back into the language.

This change adds a new rule, which makes the movability of a
pointer_element_access on a fixed size buffer dependent on the
movability of its receiver. This should require the use of a `fixed`
statement to prevent safety holes.

Fixes #28476

Merge dev15.9.x to master

Fix tests to be only conditionally skipped

Clean up and align targets with RepoToolset

Fix folder names

Fix the folder names in the myget publishing manifest

The xUnit framework appears to be reading the Skip property defined on
`ConditionalFactAttribute` vs. the one on `FactAttribute`. Even with an
unconditional throw this was interpreted as "not skipped" and hence all
tests ran.

When conditionally skipping tests for CoreClr I mistakenly
unconditionally skipped them in my attempt to document the reason for
the skip. The use of the Skip property causes the test to be
unconditionally skipped, not documenting the reason.

In addition to fixing this I prevented this from happening in the
future. It's very rare for a `[ConditionalFact]` to actually want to use
Skip as it means unconditionally skip. This is generally only for short
term bug work arounds. Hence I changed `ConditionalFactAttribute` as
follows:

- Added a new Skip property that is obsolete and has no setter.
Completely prevents the code `[ConditionalFact(..., Skip = )`
- Added a Reason property documentation in the case the test is skipped.
- Added an AlwaysSkip property in the case you always want to skip.

Misc infrastructure fixes