lib/math32: Avoid __uint128_t casts for LDC ImportC

[ML-dev-crypto]

Summary

Building the hello_d example with LDC ImportC fails because ImportC does not correctly handle direct __uint128_t C-style casts such as (__uint128_t)a and (__uint128_t)1 used in include/nuttx/lib/math32.h.

These expressions compile successfully with the existing implementation, but ImportC fails while parsing them, preventing the D example from building.

Rewrite the affected expressions to initialize __uint128_t variables first and then perform the arithmetic. This preserves the existing behavior while allowing the code to compile successfully with LDC ImportC.

Impact

Rewrite the two direct __uint128_t cast expressions in math32.h to use initialized __uint128_t variables before the arithmetic operations.

The computation remains unchanged while avoiding the current ImportC parser limitation.

Testing

Host: WSL2 Ubuntu 24.04 x86_64

Configuration: sim:nsh

Enabled CONFIG_EXAMPLES_HELLO_D.

Before this change, the build failed while compiling include/nuttx/lib/math32.h with LDC ImportC.

After applying this change, the project built successfully.

Verified the application from NSH:

nsh> hello_d
Hello World, [skylake]!
hello_d_main: Saying hello from the dynamically constructed instance
HelloWorld: mSecret=1630678024
DHelloWorld.HelloWorld: CONSTRUCTION FAILED!
Constructor
hello_d_main: Saying hello from the instance constructed on the stack
HelloWorld: mSecret=42
DHelloWorld.HelloWorld: Hello, World!!
Destructor

[github-actions]

MemBrowse Memory Report

No memory changes detected for:

• arduino-mega2560
• esp32-devkitc
• hifive1-revb
• mirtoo
• qemu-armv8a
• qemu-intel64
• rx65n-rsk2mb
• s698pm-dkit
• stm32-nucleo-f103rb