Fix packaging for the browser

[Anipik]

Master PR #40814
Related Issue #40813

@ericstj i think we dont require an approval for this one as it fixes the broken packaging ?

[Dotnet-GitSync-Bot]

I couldn't figure out the best area label to add to this PR. If you have write-permissions please help me learn by adding exactly one area label.

[ericstj]

@ericstj i think we dont require an approval for this one as it fixes the broken packaging ?

It's just in M2 approval mode now, so we just need support of me and @danmosemsft. I support this change for RC.

The only product-visible portion of this change is that it adds a PNSE to System.Drawing.Common for browser. Previously if you used this package on browser, you'd compile against a 5.0.0.0 versioned ref, but then get a 4.0.0.1 lib at runtime. Typically our packaging testing infra would have caught this, but it looks like we arent testing for browser. It might make sense to add browser to the set of RIDs we validate for.

[ghost]

Tagging subscribers to this area: @safern, @ViktorHofer
See info in area-owners.md if you want to be subscribed.

[ericstj]

I opened a PR to run package tests on browser-wasm here which should catch if we have any other instances of this: #41041

[danmoseley]

pasting from @Anipik

Description

The System.Drawing.Package is missing a ridless configuration, which results in runtime error while consuming the package on a platform which doesnt derive from windows or unix eg browser

Customer Impact

The package will work correctly for browser and non windows now unix platforms.

Regression?

No

Risk

low. This change only affects non-windows non-unix platforms.

Test changes in this PR

Package contains the netcoreapp3.0 configuration. @eric St. John is adding this scenario to package testing as well.

[danmoseley]

Approved for 5.0