[RSDK-10385] Windows build system improvements and rust_utils workarounds

[acmorrow]

• Ensure __cplusplus is set right on Windows builds.
• Don't download or link to viam_rust_utils on Windows.
• Redirect the viam_rust_utils entry points to abort on Windows.
• Adjust viamapi library name on Windows and apply /WHOLEARCHIVE
• Use std::invoke_result_t instead of result_of in C++17, where the latter is deprecated
• Add windows headers where needed.

[acmorrow]

I don't love this. I'm surprised the gRPC headers don't handle this for themselves, but without this this file doesn't compile.

The right way to do this would be with a global config precompiled header that got pulled into every .cpp we have, but I don't want to spend time on that right now, so I will file a follow-up ticket.

[acmorrow]

Filed https://viam.atlassian.net/browse/RSDK-10389

[lia-viam]

instead of doing an #if WIN32 in the cpp file, what if we did this at the CMake level:

if (NOT WIN32)
  target_link_libraries(viamsdk
    PRIVATE viam_rust_utils
  )
else()
    target_sources(viamsdk PRIVATE rust_utils_stubs.cpp)
endif()

[acmorrow]

Sure, I'll give that a try, and as long as it works (which I expect it will) I'll merge with that.

[lia-viam]

I think this looks fine assuming the main tests pass. Conan is failing but it's only as broken as it already was on main (which seems to have gotten worse recently) so not a blocker for merging.

Probably want to look into github CI for the conan package at some point however