C library: fix use of va_list for AARCH64

[tautschnig]

This is a fixup to "C model library: Support ARM64 va_list types" that

1. only changed only a subset of the code locations that required change
   and
2. did so with a crude workaround.

Really, we need to abide by ARM's procedure call standard for ARM64,
which mandates that va_list be a particular struct. See

https://github.com/ARM-software/abi-aa/blob/main/aapcs64/aapcs64.rst#definition-of-va-list
and
https://github.com/ARM-software/abi-aa/blob/main/aapcs64/aapcs64.rst#the-va-arg-macro

While at it, also add the necessary define for Visual Studio's support
of ARM64.

Fixes: #8357

• Each commit message has a non-empty body, explaining why the change was made.
• n/a Methods or procedures I have added are documented, following the guidelines provided in CODING_STANDARD.md.
• n/a The feature or user visible behaviour I have added or modified has been documented in the User Guide in doc/cprover-manual/
• Regression or unit tests are included, or existing tests cover the modified code (in this case I have detailed which ones those are in the commit message).
• n/a My commit message includes data points confirming performance improvements (if claimed).
• My PR is restricted to a single feature or bugfix.
• n/a White-space or formatting changes outside the feature-related changed lines are in commits of their own.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 78.36%. Comparing base (54c20cd) to head (20e7cef).

Additional details and impacted files

@@           Coverage Diff            @@
##           develop    #8366   +/-   ##
========================================
  Coverage    78.36%   78.36%           
========================================
  Files         1726     1726           
  Lines       188382   188382           
  Branches     18240    18240           
========================================
  Hits        147629   147629           
  Misses       40753    40753           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[kroening]

Is this something that ought to get an #ifdef ... architecture split?

[tautschnig]

Is this something that ought to get an #ifdef ... architecture split?

Are you suggesting this for improved code clarity or for other reasons?

[kroening]

The fact that *(void **)&args works on the other architectures seems like a coincidence to me.

[tautschnig]

The fact that *(void **)&args works on the other architectures seems like a coincidence to me.

This is now fixed: we no longer use that workaround and instead have suitable code depending on arch ifdefs.