Fix flags for aarch64 containers using FFTW

[cliffburdick]

No description provided.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[greptile-apps]

Greptile Summary

Conditionally configures FFTW builds based on CPU architecture, omitting x86-specific SIMD flags (--enable-sse2, --enable-avx2, --enable-avx512) on aarch64 platforms. Also adds python3-numpy to system packages.

Changes:

• Added python3-numpy to OS packages list
• Introduced architecture-aware FFTW configuration that checks cpu_target variable
• x86_64 builds use SSE2/AVX2/AVX512 optimizations (existing behavior)
• aarch64 builds omit x86-specific SIMD flags (new behavior)
• Both architectures retain OpenMP, shared library, and float/double configurations

Confidence Score: 5/5

• This PR is safe to merge with minimal risk
• The changes correctly fix a known issue where x86-specific SIMD instruction flags would fail on ARM64/aarch64 platforms. The conditional logic is straightforward, mirrors existing patterns in the codebase (lines 14-19), and the addition of python3-numpy is a benign dependency addition.
• No files require special attention

Important Files Changed

Filename	Overview
.devcontainer/recipe.py	Added architecture-specific FFTW configure flags and python3-numpy package. Conditionally omits SSE2/AVX2/AVX512 flags on aarch64.

_{Last reviewed commit: a32565b}

[greptile-apps]

_{1 file reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}