Add option to label limits and mean using secondary y axis

[francisbarton]

• [x] added unit tests
• [x] checked code coverage with covr::report() (should aim for 100%)
• [x] ran devtools::document()
• [x] ran lintr::lint_package()
• [ ] resolved all lint warnings and notes
• [x] ran styler::style_pkg() to make sure code matches the style guidelines
• [x] ran R-CMD CHECK and resolved all issues

This submission adds a label_limits Boolean flag to ptd_create_ggplot(). When turned on, this flag will add a secondary y axis to the plot, with labels that indicate the values of the UCL, LCL and mean. This is in response to issue #141 .

Suggestions for better names for the flag are welcomed.

The fork currently fails against 1 lintr criterion: cyclomatic complexity. By adding the new flag, another set of option combinations has now been added to the ptd_create_ggplot() function, which gives it a complexity value of 23, with the current linter limit set to 20.

Further changes included in this PR:

• the DESCRIPTION file now acknowledges the dependency on R >= 4.1 (this is due to the use of the base pipe, which was already in the code) - see discussion on the NHS-R Slack channel
• a package documentation file has been added using usethis::use_package_doc()
• infix namespacing functions (e.g. dplyr::) have now been added to most of the external function calls in the package, with just a few auxiliary/embedded functions (e.g. across(), aes()) being explicitly imported in the package documentation file. (Some overlap with the work in PR #180)
• the column name mean in ptd_spc_standard() has been replaced with the name mean_col (see issue #187). Referring lines of code as well as test snapshots have been updated to use the new name
• a couple of tests have been added for the new flag, and it's included in the ptd options validation function
• the point_type calculation logic in ptd_calculate_point_type() now specifies special_cause_flag == 0 rather than !special_cause_flag, which I think had been raising an error/bug somewhere.
• rlang::quo_name() calls (a superseded function) have been replaced with rlang::as_name()

I have also done some code tidying:

• the params in ptd_create_plotly were nearly all documented with the same text as the params in ptd_create_ggplot, so I replaced these with an @inheritParams line for easier maintainability
• I have standardised on UK spellings e.g. centre and colour where possible (however, plotly code uses parameters that are US spelling only)
• I have standardised on lower case letters in hex codes
• I have reduced the line length criterion in the .lintr file to 92 characters, and edited most longer lines to meet this. (I prefer an 80 character line maximum but I thought 92 was at least better/tidier than 120 chars)

This will also close issue #187 if accepted.

[tomjemmett]

thanks! I think we need to merge #180 and #182 first. These are probably going to cause some merge conflicts by looking at the comparison's of the branches - hopefully these should be too problematic!

[francisbarton]

Thanks Tom. I think most of the issues will be 'stylistic' like linting, line length, and the use of dplyr:: etc prefixes. Possibly an issue around adoption of the base pipe.

[tomjemmett]

I've attempted to resolve conflicts. If they fail will revert my commit and have another look!

[tomjemmett]

going to merge and apply some fixes in another PR. many thanks @francisbarton !