Write doubles with the shortest digits (Żmij), digits in registers (#5634)

* Write doubles with the shortest digits (Żmij), digits in registers

Write doubles with the conversion of Zmij by Victor Zverovich (MIT),
ported to C++11 (detail/conversions/zmij.hpp). It finds the
shortest decimal that reads back as the same double, and the
closest one if there are several. Grisu2, used until now, is fast
but not always shortest: it sometimes writes a 17th digit where 16
suffice, or a last digit that is not the closest. The layout is
unchanged (1.5, 100.0, 1e+100, -0.0); float keeps Grisu2.

Digits are converted eight at a time with the BCD conversion of
Xiang JunBo, as in Zmij, and written with one byte swap per eight
digits and fixed-size moves instead of per-digit loops. Leading and
trailing zeros are counted from those bytes. to_chars() uses a
local buffer when the caller's is shorter than the 41 bytes this
may write. The powers of ten come from the number-parsing table,
adjusted where it holds values rounded up, and extended with Zmij's
compressed tables beyond 10^308.

write_shortest() converts its 16 digits in one vector register
(SSE2 on x86-64, NEON on 64-bit Arm, both baseline) and inserts the
decimal point inside the register, avoiding a store-forwarding
stall that cost about 25% of the time to write a double. dump()
writes floats and integers straight into the serializer's write
buffer instead of copying them from a member buffer, and small
integers eight digits at a time. read_eight_bytes() and
parse_eight_digits() are marked always-inline, which GCC had been
calling out of line in the number-parsing loops.

Of one million random doubles, about 0.14% are now written with
different digits, always to a value that still reads back as the
same double.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

* Fix CI findings in the Zmij writer and its tests

- bit_ops.hpp: include macro_scope.hpp (JSON_HEDLEY_ALWAYS_INLINE); fixes IWYU
- to_chars.hpp: C4100 for the unused parameter in release builds, clang-tidy
  sign comparison, cpplint runtime/int, GCC -Wstrict-overflow (unsigned abs)
- unit-to_chars.cpp: parse with the library instead of strtod (MinGW's strtod
  rounds some 16/17 digit inputs wrongly); no floating-point std::to_chars
  with icpc

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

* Select the Zmij conversion by a trait, not by the double overload

The non-template overloads for double asserted binary64 doubles wherever json.hpp was included, so the library no longer compiled where double is not IEEE 754 binary64 (AVR, -fshort-double). A trait now picks Zmij for any binary64 type, including a long double of that format (MSVC, Apple Arm), and Grisu2 for the others. Remove the unused write_short_decimal, powers_of_ten_16, zmij::decimal, zmij::to_decimal, and shortest_digits(double).

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

* Credit Zmij's authors in to_chars.hpp, the README, and the license page

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

* Silence MSVC C4127 for a platform-constant check in the to_chars test

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

* Regenerate the amalgamated headers

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

---------

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

include/nlohmann/detail/bit_ops.hpp

@@ -14,7 +14,7 @@
     #include <intrin0.h> // __umulh, _umul128, _BitScanForward64, _BitScanReverse64
 #endif
 
-#include <nlohmann/detail/abi_macros.hpp>
+#include <nlohmann/detail/macro_scope.hpp> // JSON_HEDLEY_ALWAYS_INLINE, NLOHMANN_JSON_NAMESPACE_BEGIN
 
 // Portable bit-level helpers for the number and string scanners. They use
 // compiler builtins or platform-specific intrinsics where available and plain
@@ -103,8 +103,9 @@ inline uint128_parts full_multiplication(std::uint64_t a, std::uint64_t b) noexc
 #endif
 }
 
-/// eight bytes as a little-endian word (a single load on little-endian targets)
-inline std::uint64_t read_eight_bytes(const unsigned char* b) noexcept
+/// eight bytes as a little-endian word (a single load on little-endian
+/// targets; always inlined, as GCC otherwise calls it in the number loops)
+JSON_HEDLEY_ALWAYS_INLINE std::uint64_t read_eight_bytes(const unsigned char* b) noexcept
 {
 #if defined(_MSC_VER) || defined(__x86_64__) || defined(__i386__) || (defined(__BYTE_ORDER__) && defined(__ORDER_LITTLE_ENDIAN__) && __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
     // the byte order already matches (all MSVC targets are little-endian)
@@ -120,7 +121,7 @@ inline std::uint64_t read_eight_bytes(const unsigned char* b) noexcept
 }
 
 /// eight bytes as a little-endian word
-inline std::uint64_t read_eight_bytes(const char* p) noexcept
+JSON_HEDLEY_ALWAYS_INLINE std::uint64_t read_eight_bytes(const char* p) noexcept
 {
     return read_eight_bytes(reinterpret_cast<const unsigned char*>(p)); // NOLINT(cppcoreguidelines-pro-type-reinterpret-cast)
 }