benchmarks/sd_bench: Fix sd_bench in CMake build system.

[Felix-LJY]

Summary

The sd_bench application under testing/sd_bench was not built when
using the CMake build system. The application only had a Makefile-
based build path, so users building NuttX with CMake/Ninja could not
find sdbench in the resulting image.

This change adds the missing CMakeLists.txt for sd_bench and registers
it with the applications CMake build system, so the application is
included in the build when CONFIG_BENCHMARK_SD_BENCH is enabled.

Functional areas affected:
benchmarks/sd_bench/CMakeLists.txt (new file)
benchmarks/sd_bench/CMakeLists.txt (registered into apps CMake)

References:
The corresponding nuttx repository PR adds N32H7 support and uses
sdbench for SDMMC performance verification. Without this fix,
sdbench could only be built with the Makefile build system.

Impact

Is new feature added? NO - Fixes an existing application
(sd_bench) that was not available under the CMake build system.

Impact on user? YES - Users building with CMake/Ninja and enabling
CONFIG_BENCHMARK_SD_BENCH will now get the sdbench command in the
final image, as they already did with the Makefile build system.

Impact on build? YES - Adds a new CMakeLists.txt for sd_bench and
registers it. No existing build configuration is modified.

Impact on hardware? NO - No hardware-specific changes.

Impact on documentation? NO - No documentation changes required.

Impact on security? NO - No security-relevant changes.

Impact on compatibility? NO - The Makefile build system is
unaffected. Existing users of the Makefile path see no change.

Testing

Build Host(s):

• OS: Ubuntu Linux
• CPU: x86_64 (AMD)
• Compiler: GCC (ARM GNU Toolchain 15+)

Target(s):

• Arch: ARM (Cortex-M7)
• Board: n32h762iil7:nsh

Verification steps:

1. Build nuttx-apps with CMake/Ninja:
   cmake -B build -DBOARD_CONFIG=n32h762iil7:nsh -G Ninja
   ninja -C build
2. Confirm the resulting image contains the sdbench command.
3. Boot on hardware and run:
   nsh> sdbench -b 4096
4. Verify the benchmark runs and reports valid read/write rates.

Testing logs after change (summary):

nsh> sdbench -b 4096
Using block size = 4096 bytes, sync = false

Testing Sequential Write Speed...
Run 1: 1321.2 KB/s
Run 2: 1788.0 KB/s
...
Avg : 1423.2 KB/s, 14.891 MB written.

Testing Sequential Read Speed...
Run 1: 4566.0 KB/s
Run 2: 4992.7 KB/s
Avg : 4728.1 KB/s, 14.891 MB and verified

PR verification Self-Check

[x] This PR introduces only one functional change.
[x] I have updated all required description fields above.
[x] My PR adheres to Contributing Guidelines and Documentation.
[ ] My PR is still work in progress (not ready for review).
[x] My PR is ready for review and can be safely merged.