switch to directly calculated growth rate and extend model options, take 2

[sbfnk]

Attempts to resurrect #307 as work in progress. Copying text from there:

This PR is meant to supersede #213 and #214. It

• finalises the work commenced in Standardises generation of infections #214 to support three different models
• merges Standardises generation of infections #214 into Update growth calculation #213 and implements most of the suggestions made there by @hsbadr
• rebases Update growth calculation #213 onto current master
• works with default options and a simple example (the README)

Things left to do:

• expose and document the model options and ensure the stan model options are set consistently
• Update the README so that it works.
• Remove backcalc_opts() from all input
• Remove rt_opts() from all input
• Move generation_time_opts() inside of process_opts() as an argument.
• Check postprocessing works as expected
• Check that all examples and docs work as expected
• Document model based on update specification
• Get passing CRAN check if not already
• more extensive testing
• update NEWS

[sbfnk]

To do for being able to revisit this:

• rebase to main
• reduce the number of models to 2 (growth rate model is the same as renewal model with fixed generation time of 1)
  and then go back to the items in the original issue

[sbfnk]

Closing as most suggestions in here have now made it into main one way or the other (and we have diverged to extract any use from this)