use native Julia code in round example

[udohjeremiah]

This !!! note section on round introduces a concept or function not native to Julia code, where a better Julia solution could have been used:

  │ Note
  │
  │  Rounding to specified digits in bases other than 2 can be
  │  inexact when operating on binary floating point numbers. For
  │  example, the Float64 value represented by 1.15 is actually less
  │  than 1.15, yet will be rounded to 1.2. For example:
  |
  │  julia> @sprintf "%.20f" x
  │  "1.14999999999999991118"

The @sprintf function introduced there complicates things for some number of reasons:

• Julia users not acquainted with the Printf module or C printf style would not get the understanding of what that part is trying to do or explain. Sure they can lookup @sprintf in the docs, but then they have to figure out what the %.20f does and all that stuffs, just to understand the example.
• Trying to replicate the example fails because it does not show where @sprintf comes from (i.e. no using Printf is shown there, so users know its exported).

Just to understand an example, and then one has to figure out all that sutff - Huh, I think its uncalled for. Why not just use Julia's big to show what's happening there, its same with the @sprintf (just the difference being that @sprintf shows the results in 20 digits after decimal point):

julia> using Printf

julia> x = 1.15
1.15

julia> big(x)
1.149999999999999911182158029987476766109466552734375

julia> @sprintf "%.20f" x
"1.14999999999999991118"