lalapps_binj build fails with clang: implicit-int-float-conversion

Compiling lalapps_binj with clang-10 (at least) causes a compiler error:

make[8]: Entering directory '/builds/duncanmmacleod/lalsuite/lalsuite-6.81.1/_build/sub/lalapps/src/power'
  CC       binj.o
  CC       power.o
../../../../../lalapps/src/power/binj.c:900:58: error: implicit conversion from 'unsigned long' to 'double' changes value from 18446744073709551615 to 18446744073709551616 [-Werror,-Wimplicit-int-float-conversion]
        sim_burst->waveform_number = floor(gsl_ran_flat(rng, 0, ULONG_MAX));
                                           ~~~~~~~~~~~~         ^~~~~~~~~
/usr/lib/llvm-10/lib/clang/10.0.1/include/limits.h:57:37: note: expanded from macro 'ULONG_MAX'
#define ULONG_MAX (__LONG_MAX__ *2UL+1UL)
                   ~~~~~~~~~~~~~~~~~^~~~
1 error generated.

See https://git.ligo.org/duncanmmacleod/lalsuite/-/jobs/1202343#L9054.