Fix: asserts unintentially being partially disabled in release-builds (#11268)

The nlohmann-json header file includes assert.h, which rudely resets
the assert macro to what that header thinks is right. As we set the
assert macro to be active with release builds when WITH_ASSERT is
active, this means that every file including nlohmann-json has their
asserts disabled (for release-builds) but files that don't do no.

Let's avoid this issue, by telling nlohmann to not include assert.h.

Remove: last remnants of SunOS (#11210)

We have no evidence SunOS has been tried in the last few years,
and the code is awkward enough to just remove.

Codechange: Base CargoArray off std::array.

This avoids needing to define array accessors and allows use of
default value initialization.

Change: include fmt.h C++ headers in stdafx.h

This to prevent compilation issues between runs with and without precompiled
headers. Also remove the headers from the rest of the code base as they are
not needed there anymore, although they do relatively little harm.

Codechange: introduce a framework for all our timers

IntervalTimer and TimeoutTimer use RAII, and can be used to replace
all the time-based timeouts, lag-detection, "execute every N" we
have.
As it uses RAII, you can safely use it as static variable, class
member, temporary variable, etc. As soon as it goes out-of-scope,
it will be safely removed.
This allows for much easier to read code when it comes to intervals.

Codechange: use "[[maybe_unused]]" instead of a wide variety of other ways we had

While at it, replace OTTD_ASSERT with WITH_ASSERT, as this
is always set if assert() is valid. No matter if NDEBUG is set
or not.

Codechange: use thread safe time functions

Functions like localtime, gmtime and asctime are not thread safe as they (might) reuse the same buffer. So use the safer _s/_r variant for localtime and gmtime, and use strftime in favour of asctime.

Fix: clang-cl build (#9018)

Remove macro redefinitions
Add final and fallthrough attributes for clang-cl

Fix: [Cygwin] Fix missing uint definition

In file included from src/settingsgen/../string_func.h:30,
from src/settingsgen/settingsgen.cpp:11:
src/settingsgen/../core/bitmath_func.hpp:34:15: error: 'uint' does not name a type; did you mean 'uint8'?
34 | static inline uint GB(const T x, const uint8 s, const uint8 n)
| ^~~~
| uint8

Codechange: [Win32] Increase SDK version for 32-bit target.

Use the same Windows XP target as for 64-bit. Current MSVC version will
not produce a binary that works on anything earlier anyway.

Revert c1fddb9, 639cfa4: "Codechange: Check if access __attribute__ is supported before trying to use it." (#8526)

This reverts commit c1fddb9a6ae5c3af6865461a7295788a341011a2 and 639cfa43d23aa142cabbf2f08ec20a2133b0503e.

access_mode "none" is only supported by GCC11, but introduced
after it branched. So there are GCC11.0s out there that do not
support it. We will have to wait for GCC11.1 to hit before we
can re-add this.

Codechange: Unconditionally use static_assert

We're well past having to support non-C++11 compliant compilers now.

Fix: Globally apply preprocessor directive coding style

Global; except for the 32-bit SSE blitter, which has some #DEFINEs
in not-very-nice places.

Fix: Silence some warnings when building with clang-cl on VS 2019

Clang-cl presents as both _MSC_VER and __clang__ in the preprocessor which makes some things confusing.

Add: introduce CMake for project management

CMake works on all our supported platforms, like MSVC, Mingw, GCC,
Clang, and many more. It allows for a single way of doing things,
so no longer we need shell scripts and vbs scripts to work on all
our supported platforms.

Additionally, CMake allows to generate project files for like MSVC,
KDevelop, etc.

This heavily reduces the lines of code we need to support multiple
platforms from a project perspective.

Addtiionally, this heavily improves our detection of libraries, etc.

Fix: [Cygwin] Fix missing definitions in stdafx

Fix the following compile errors:

In file included from openttd/src/debug.h:15:0,
from openttd/src/string.cpp:13:
openttd/src/cpu.h:35:19: error: 'uint' was not declared in this scope
bool HasCPUIDFlag(uint type, uint index, uint bit);

openttd/src/string.cpp: In function 'char* strcasestr(const char*, const char*)':
openttd/src/string.cpp:548:7: error: 'strncasecmp' was not declared in this scope
if (strncasecmp(haystack, needle, needle_len) == 0) return const_cast<char *>(haystack);

openttd/src/strgen/strgen_base.cpp: In function 'void EmitPlural(Buffer*, char*, int)':
openttd/src/core/alloc_func.hpp:136:6: error: 'alloca' was not declared in this scope
(T*)alloca((num_elements) * sizeof(T)))

Signed-off-by: Joe Stringer <joe@wand.net.nz>

Remove: DOS support

In 10 years there was no active development on DOS. Although it
turned out to still work, the FPS was very bad. There is little
interest in the current community to look into this.

Further more, we like to switch to c++11 functions for threads,
which are not implemented by DJGPP, the only current compiler
for DOS.

Additionally, DOS is the only platform which does not support
networking. It is the reason we have tons of #ifdefs to support
disabling networking.

By removing DOS support, we can both use c++11 functions for threads,
and remove all the code related to disabling network. Sadly, this
means we have to see DOS go.

Of course, if you feel up for the task, simply revert this commit,
and implement stub c++11 functions for threads and stub functions
for networking. We are more than happy to accept such Pull Request.

Fix: warnings when compiling for a recent version of Haiku

This might break older Haiku versions, but it is hard to tell.

Remove: BeOS support (deprecated by Haiku)

In 10 years there is no commit to change how BeOS works, and we
have no active maintainer for it. It is unlikely it works in its
current state (but not impossible).

With the arrival of SDL2 (and removal of SDL), BeOS is no longer
support. SDL2 suggests to use Haiku instead of BeOS.

Remove: MorphOS / AmigaOS support

In 10 years there is no commit to change how MorphOS works, and we
have no active maintainer for it. It is unlikely it works in its
current state (but not impossible).

With the arrival of SDL2 (and removal of SDL), MorphOS is no longer
support. There is an SDL2 port for MorphOS, but it is not maintained
by upstream SDL2, and nobody can currently test it out.

If anyone wants to re-add MorphOS, please do (revert this patch,
fix the problems, and create a Pull Request). If you need any help
doing so, let us know! It is not that we don't like MorphOS, it is
that we don't have anyone fixing the problems :(

Change: use vcpkg integration in Visual Studio

Removes the need for us to provide OpenTTD-useful package