multipleOf contains incorrectly optional (bignum dependent) tests

I'm working on updating my validator to consume the latest version of the test suite (after about 18 months; I usually aim for once per year but hey...), and hit issues with the multipleOf.json tests now containing bignum scenarios (see #438). I also noticed that there's recent discussion about changing or adding more bignum tests to multipleOf (see #534).

As @Julian called out in #438:

Unless, is arbitrary-precision support optional?

Arbitrary precision is optional yeah. You are allowed to use IEEE floats or doubles (and presumably also this -- 16 byte floats)

Can we agree to place these tests in the optional suite so that implementors who wish to leverage the official suite but don't have bignum support can more easily navigate around this obstacle? In #438, the float version of the test was placed under optional, but the integer version was not.

/cc'ing some folks who've been involved in the other issues: @karenetheridge, @Zac-HD

Start by inspecting the multipleOf.json tests and the optional suite, then compare the integer and float bignum scenarios discussed in issues #438 and #534. Done means bignum-dependent multipleOf tests are consistently placed in the optional suite so implementations without arbitrary-precision support can skip them.