new 'float' math modules

[Bruno Haible]

I wrote:
> For 22 <math.h> functions that ought to take 'float' arguments and
> return a 'float' result, MSVC 9 only provides a macro definition.
> This is not POSIX compliant, see POSIX:2008 / System Interfaces /
> General Information / Use and Implementation of Interfaces
> <http://pubs.opengroup.org/onlinepubs/9699919799/functions/V2_chap02.html>
> section 2.1.1, point 2.

I've finished adding gnulib modules for each of these:
  fabsf
  fmodf
  modff
  frexpf
  ldexpf
  sqrtf
  expf
  logf
  log10f
  powf
  sinf
  cosf
  tanf
  asinf
  acosf
  atanf
  atan2f
  sinhf
  coshf
  tanhf

Bruno
-- 
In memoriam Engin Ceber <http://en.wikipedia.org/wiki/Engin_Ceber>