Add functionality for simple validations (e.g., df %>% col_vals_gt(...))

[rich-iannone]

The idea is to pass a data object directly to a validation function and get a re-usable output (e.g., vector of logical values) that can be used in other functions. This would be very useful for joint validations where we could have:

df %>% <validation_function>(...) & 
df %>% <validation_function>(...)

And the resultant vector of logicals could show which rows jointly passed (of course, one has to ensure that the input is passed to the validation functions unchanged).

This shouldn't affect the existing API that much. The first argument of any validation function will change from agent to ... where each function will internally sort out whether to use an agent object or immediately interrogate. The ... will also be useful if we decide to wrap inputs in some helper function (e.g., jointly(), etc.).

[rich-iannone]

This has been added. The question that remains is whether the output is useful enough. A new issue will be opened to address that.