Some to do items before writing paper

[cianwilson]

• try increasing Poisson 2D parallel grid size to see if it does any better on github workflows (Batchelor looks great and is a bigger problem)
• check residual markdown in Blankenbach for consistent use of test space notation
• comment on tf in demo cases
• compare time dependent and steady state like in figure in paper
• add locations map and more intro to global suites

[cianwilson]

• use help on global suite instead of ?

[cianwilson]

• split up sz themes and variations?

[cianwilson]

• test eta projection in parallel