compiler: add HAVE_EXPRESSION_STATEMENT macro for gcc

[GUIDINGLI]

Summary

compiler: add HAVE_EXPRESSION_STATEMENT macro for gcc

this will open fast version of div_const when meet 64/32 when use gcc compiling

Impact

div_const in
include/nuttx/lib/math32.h

Testing

CI

[nuttxpr]

[Experimental Bot, please feedback here]

No, this PR does not fully meet the NuttX requirements as described. While it provides a summary of the change, it lacks crucial details. Here's a breakdown:

Missing Information:

• Summary:

  • Why is this change necessary? What problem does it solve? Is the current division slow? Are there benchmarks showing the improvement? Just stating it "opens a fast version" isn't sufficient.
  • How exactly does it work? What is HAVE_EXPRESSION_STATEMENT and how does its presence affect div_const? What code paths are changed? Provide a snippet if possible.
  • Related Issues: Are there any related NuttX or NuttX Apps issues this PR addresses?

• Impact: The impact section is severely lacking. It only mentions the affected file. Address all the impact points:

  • New/Changed Feature? Is this an optimization of an existing feature or a new one?
  • Impact on user, build, hardware, documentation, security, compatibility: Even if the answer is "NO" for each, explicitly state it. If there is an impact, describe it. For example, the build impact is likely YES as a new macro is introduced.

• Testing:

  • Build Host(s): Be specific. "CI" isn't enough. List the actual CI environments used (e.g., "GitHub Actions: Ubuntu Latest, macOS Latest").
  • Target(s): What architectures and boards were tested? "sim", "RISC-V", "ARM" are too broad. Specify the specific simulator, RISC-V board, and ARM board.
  • Logs: Providing "before" and "after" logs is crucial. These should demonstrate the performance improvement or the corrected behavior. Empty code blocks are unacceptable.

Example of Improved Summary & Impact:

## Summary

This PR introduces the `HAVE_EXPRESSION_STATEMENT` macro to enable a faster version of `div_const` when compiling with GCC.  Currently, the 64/32 division is inefficient [link to benchmark/profiling data if available]. This macro allows the compiler to optimize the division operation, resulting in a performance improvement [quantify the improvement if possible]. This change modifies the `div_const` function in `include/nuttx/lib/math32.h`.

## Impact

* Is new feature added? NO (optimization of existing feature)
* Is existing feature changed? YES (`div_const` performance improved)
* Impact on user: NO
* Impact on build: YES (new macro `HAVE_EXPRESSION_STATEMENT` needs to be handled by the build system)
* Impact on hardware: NO
* Impact on documentation: YES (documentation should be updated to explain the new macro and its impact on `div_const`)
* Impact on security: NO
* Impact on compatibility: NO
* Anything else to consider:  [mention anything relevant, e.g., potential impact on code size]

By providing the missing details and explicitly addressing each point in the template, the PR will be much clearer and easier to review, ultimately increasing the chances of it being accepted.

[xiaoxiang781216]

why need slow version