update vignette

[pearsonca]

• [ ] split: non-identity in "basic" vignette, with identity in "intermediate"
• [ ] basic vignette should focus on add-hbm-functions to your model => basics of how to assess consequence of adding them
• [ ] intermediate on use them for a more complex model + more comparisons
• [ ] switch to using data-raw => inst to precompute simulations pertinent to the vignette

[pearsonca]

written, works, shows effects (-ish? bunch of non-epidemics could potentially be pruned - is that fair play?)

• [ ] slow - reasonable to do pure binomial simulation version of matching? that's a plausible non-cheat in compartmental land, but i think neglects real individual uncertainty. that's perhaps a useful element to highlight?
• [ ] probably going to choke on CRAN process, since using packages in the vignette build that aren't required for package build. legal for those to just go in suggests?

[seabbs]

yup legal to put packages used in vignettes into suggests and can pre-compile the vignettes to avoid slow run times