drivers/mtd: keep MTDIOC_ERASESECTORS inside its partition

[royzah]

Note: Please adhere to Contributing Guidelines.

Summary

MTDIOC_ERASESECTORS on a partition went straight to the parent with no bound, so an erase from the last block on hit the next partition. part_erase() already checks it.

Call part_erase(). Plus the nxstyle fixes in the file.

Impact

MTD partitions. An erase past the end of a partition fails instead of erasing the next one.

Testing

Host: Ubuntu 24.04, x86_64. Builds and QEMU runs in ghcr.io/apache/nuttx/apache-nuttx-ci-linux (QEMU 6.2, Arm GNU GCC 13.2, xPack RISC-V GCC 14.3). Hardware: i.MX93 (Cortex-A55), PX4 kernel build.

i.MX93, 64 MB FlexSPI NOR in partitions, MTDIOC_ERASESECTORS from the block past the end of a 7680-block partition:

	result
before	erases the next partition's first block (read from the code, not run: it holds an image)
after	-ENXIO

erase block 7680 of a 7680-block partition: -1, errno 6

i.MX93, PX4 kernel build, this series applied: tests isolation

INFO  [tests] RUNNING TEST: test_erase_bounded
INFO  [tests] erase block 7680 of a 7680-block partition: -1, errno 6
INFO  [tests] TEST PASSED: test_erase_bounded
INFO  [tests]   Tests passed :      15
INFO  [tests]   Tests failed :      0

tools/checkpatch.sh clean.

[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

[royzah]

Two approvals and CI is green, could a committer merge this one when there's a moment? Thanks!

[xiaoxiang781216]

Two approvals and CI is green, could a committer merge this one when there's a moment? Thanks!

but need two committer/pmc approved.