Update support/testing for overintegration

[MTCam]

Meant to replace #1069. This change set adds tests from the Mengaldo paper on dealiasing to verify and exercise overintegration in mirgecom. The following are added or modified:

• Examples

  • agitator.py: The variable coefficient advection setup from Mengaldo Sec. 4.1
  • flow-over-cylinder.py: The subsonic inviscid flow over cylinder from Mengaldo Sec. 4.3.1

• Tests

  • test_operators.py: Quadrature verification with Chebyshev polynomials
  • test_overintegration.py: Variable coefficient overintegration verification from Mengaldo Sec. 4.1, Figure 7
  • overintegration (and tpe) versions added to most flow, and operator tests as appropriate

• Support

  • discretization.py: Fixed up choosing the proper order argument to quadrature group based on element type
  • inviscid.py: Added central numerical flux for divergence of inviscid flux

• Some support and fixes required (mostly upstream pkgs)

  • overintegration not properly used in WADG (illinois-ceesd/grudge@7c8eaec)
  • TPE normals not on quadrature domain when using overintegration in mirgecom (fixed upstream)
  • Revert TPE to use simplex operator chain; fixed TPE/overintegration failing operator/flow tests in mirgecom
  • grudge integration utility uses mass matrix instead of direct quadrature (illinois-ceesd/grudge@2e54482)
  • mirgecom discretization creation utility does not properly create quadrature domain of correct order for TPE (this PR)
  • transform chain breaks when setting quadrature order < 2p (illinois-ceesd/loopy@fc274fa)
  • lazy required for quantitative test from Mengaldo to run in pytest in reasonable time (this PR)
  • undo actx switch for TPEs in gradient testing (this PR)

Questions for the review:

• Is the scope and purpose of the PR clear?
  • The PR should have a description.
  • The PR should have a guide if needed (e.g., an ordering).
• Is every top-level method and class documented? Are things that should be documented actually so?
• Is the interface understandable? (I.e. can someone figure out what stuff does?) Is it well-defined?
• Does the implementation do what the docstring claims?
• Is everything that is implemented covered by tests?
• Do you see any immediate risks or performance disadvantages with the design? Example: what do interface normals attach to?

[anderson2981]

Yes!