Toolchain: Enable float FFTW linking for building with MKL

[Growl1234]

MKL supports both single and double precision at the same time, so adding -DENABLE_FLOAT_FFTW=ON flag is of no harm. It wasn't added to GNU toolchain building script in case some people link the FFTW package installed by their own (it's easy via simply passing --with-fftw=system to the end of ./toolchain_gnu.sh command) with no single precision support (its single/double precision does not share the same configuration and cannot be compiled at once).

[Copilot]

Pull request overview

This PR updates MKL-based toolchain build scripts to always enable ABACUS’ single-precision FFTW code paths when configuring with CMake, aligning with the intent that MKL can support both single and double precision FFT interfaces.

Changes:

• Enable -DENABLE_FLOAT_FFTW=ON in the Intel (icpx/mpiicpx) toolchain build script.
• Enable -DENABLE_FLOAT_FFTW=ON in the GCC+MKL toolchain build script.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
toolchain/build_abacus_intel.sh	Adds -DENABLE_FLOAT_FFTW=ON to the CMake configure command for Intel+MKL builds.
toolchain/build_abacus_gcc-mkl.sh	Adds -DENABLE_FLOAT_FFTW=ON to the CMake configure command for GCC+MKL builds.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.