Generate FSQRT from calls to the sqrt function, which

allows appropriate backends to generate a sqrt instruction.

On x86, this isn't done at -O0 because we go through
FastISel instead.  This is a behavior change from before
this series of sqrt patches started.  I think this is OK
considering that compile speed is most important at -O0, but
could be convinced otherwise.

llvm-svn: 82778