drop warn once functions

[a-kramer]

The "warn once" functions typically warn about correct behavior. Example:

library(errors)
x <- set_errors(9,1)
x + 1

Warning message:
In 'Ops' : non-'errors' operand automatically coerced to an 'errors' object with no uncertainty 

I think these functions can be removed without substitute. It's overly verbose, and a bad first impression when trying things out.

It was a good move to warn about these things only once, but even that is unnecessary in my opinion.

[Enchufa2]

What do you mean by correct? If it wasn't correct, it would be an error, right? :)

I may consider dropping the coercion warning if we decide that this coercion is proper. Among other reasons, because units does this already. But what about this one:

set_errors(1:10, 0.1) > set_errors(1:10, 0.2)
#> Warning: In '>' : boolean operators not defined for 'errors' objects,
#> uncertainty dropped
#>  [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE

I don't think it's a good idea to drop this one, because comparison between numbers with uncertainty is not well defined, and we drop the uncertainty for convenience.