Add NoAutoDiff

[oschulz]

Checklist

• Appropriate tests were added
• Any code changes were done in a way that does not break public API
• All documentation related to code changes were updated
• The new code follows the
  contributor guidelines, in particular the SciML Style Guide and
  COLPRAC.
• Any new documentation only uses public API

Additional context

Add any other context about the problem here.

[oschulz]

Closes #121

[oschulz]

With this PR, we get

julia> DifferentiationInterface.gradient(sum, NoAutoDiff(), rand(5))
ERROR: NoAutoDiffAvailable("Automatic differentiation can not be used with NoAutoDiff()")
Stacktrace:
 [1] mode(::NoAutoDiff)
   @ ADTypes /user/.julia/dev/ADTypes/src/dense.jl:543
 [2] pullback_performance(backend::NoAutoDiff)
   @ DifferentiationInterface /user/.julia/dev/DifferentiationInterface/DifferentiationInterface/src/utils/traits.jl:102
[...]

[oschulz]

@gdalle I was using

struct NoAutoDiffSelectedError <: Exception
    msg::AbstractString
end

instead of msg::String because Base.ArgumentError, Base.ErrorException and similar are defined like that. I don't know if that's to avoid memory allocation due to string conversion when an exception object is created?

[gdalle]

No clue. I guess we can leave it like this but if it turns out to be a problem we'll switch back