MSVC: Treat warnings as errors

[chrisdembia]

Brief summary of changes

This PR causes MSVC to treat warnings as errors and checks an additional warning about signed/unsigned integers for consistency with warnings generated by Clang.

I'm making this change to help with a similar change in Moco.

Testing I've completed

Ran all ctests.

CHANGELOG.md (choose one)

• no need to update because...related to build system.

---

This change is 

[aymanhab]

@chrisdembia Most of these are due to mixing int and size_t, can we use size_t instead of casting whenever possible? Thanks

[chrisdembia]

A while ago, I was trying to figure out the best way to handle the different integer types in loops, and my first thought was to use size_t. But that just moved the warning elsewhere; we can't avoid the cast because we use int throughout Simbody's numeric types, etc. For example, we would need to use simbodyVec[(int)index)] in many places.

So, using int is consistent with Simbody (e.g., https://github.com/simbody/simbody/blob/c6f465bc7de0ca2adaecdbffac6076a173f5a347/Simbody/src/MultibodySystemRep.h#L393).

I also referred to the C++ Core Guidelines: https://github.com/isocpp/CppCoreGuidelines/blob/master/CppCoreGuidelines.md#es107-dont-use-unsigned-for-subscripts-prefer-gslindex.

These guidelines suggest using a different type from a third-party library gsl::index. This is the explanation they give:

The built-in array uses signed subscripts. The standard-library containers use unsigned subscripts. Thus, no perfect and fully compatible solution is possible (unless and until the standard-library containers change to use signed subscripts someday in the future). Given the known problems with unsigned and signed/unsigned mixtures, better stick to (signed) integers of a sufficient size, which is guaranteed by gsl::index.

They suggest sticking to signed integers. I know that int may not be of sufficient size but I'm not sure that's a practical issue for us, considering that Simbody uses int everywhere anyway.

[aymanhab]

Are these changes related?

[aymanhab]

Reviewed 15 of 15 files at r1.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @chrisdembia)