[NRBF] skip the most time-consuming test case for non-Release builds to avoid timeouts for checked builds

[adamsitnik]

I was not able to reproduce a timeout, but I was able to notice that it takes a LOT of time to run (more than 10 minutes on a really beefy PC). I've attached a profiler and found the test case which was executing the worst case scenario (non-seekable stream from which we read 512k bytes one by one).

The command I've used after building the repo:

cmd.exe /C start /affinity F "D:\projects\runtime\artifacts\bin\testhost\net10.0-windows-Debug-x86\dotnet.exe" exec --runtimeconfig System.Formats.Nrbf.Tests.runtimeconfig.json --depsfile System.Formats.Nrbf.Tests.deps.json C:\Users\adsitnik\.nuget\packages\microsoft.dotnet.xunitconsolerunner\2.9.2-beta.24605.1\build\..\tools\net\xunit.console.dll System.Formats.Nrbf.Tests.dll -xml testResults.xml -nologo -notrait category=OuterLoop -notrait category=failing

  Discovering: System.Formats.Nrbf.Tests (method display = ClassAndMethod, method display options = None)
  Discovered:  System.Formats.Nrbf.Tests (found 150 test cases)
  Starting:    System.Formats.Nrbf.Tests (parallel test collections = on [4 threads], stop on fail = off)
    System.Formats.Nrbf.Tests.EdgeCaseTests.CanReadArrayOfAnySize [SKIP]
      Condition(s) not met: "Is64BitProcess"

fixes #110285

[jkotas]

Is 512_001 particularly interesting value to test for some reason?

Would it be simpler to just change 512_001 to something smaller, like 20_001?

[adamsitnik]

Nothing in particular, I am going to apply your suggestion.

[jkotas]

Thanks