Switch elliptic DG to a primal formulation

[nilsvu]

Proposed changes

Handle auxiliary boundary corrections through a weak divergence rather than a Schur complement. This changes how Jacobians are applied on boundaries. It turns out this primal formulation is more stable when the Jacobians are large (for large outer boundaries).

Upgrade instructions

Code review checklist

• The code is documented and the documentation renders correctly. Run
  make doc to generate the documentation locally into BUILD_DIR/docs/html.
  Then open index.html.
• The code follows the stylistic and code quality guidelines listed in the
  code review guide.
• The PR lists upgrade instructions and is labeled bugfix or
  new feature if appropriate.

Further comments

[nilsvu]

@nilsdeppe this is the next PR in the series to fix the large outer boundaries. Can you review this one too?

[nilsdeppe]

Yep!