fix(FSCommon): add nRF52 LittleFS empty-name guard to getFiles()

[Copilot]

getFiles() iterated with while (file) without the file.name()[0] guard that listDir() already uses to work around the Adafruit LittleFS nRF52 glue bug (issue 4395), where openNextFile() can return a File with an empty name. This could cause empty-name entries to be pushed into results and the loop to spin until maxCount was exhausted.

Changes

• src/FSCommon.cpp — updated the loop condition in getFiles() to match listDir():

// Before
while (file) {

// After
while (file && file.name()[0]) { // file.name()[0] check is a workaround for a bug in the Adafruit LittleFS nrf52 glue (see issue 4395)

🤝 Attestations

• I have tested that my proposed changes behave as described.
• I have tested that my proposed changes do not cause any obvious regressions on the following devices:
  • Heltec (Lora32) V3
  • LilyGo T-Deck
  • LilyGo T-Beam
  • RAK WisBlock 4631
  • Seeed Studio T-1000E tracker card
  • Other (please specify below)

[thebentern]

@copilot trunk fmt

[Copilot]

@copilot trunk fmt

Done — ran clang-format on src/FSCommon.cpp in 3dc8ab4.

[shalberd]

Hi all, I know I repeat myself, but yet: again a big thank you to you all for your dedication in getting bugs resolved, nodes running more stable, and so on.

I had a few mountain nrf52 Seeed Solar P1 nodes running in firmware 2.7.11, curiosity ... in any case, they stopped forwarding messages after about 5 months. It wasn't a node failure itself, it was a filesystem error most likely. After I updated to firmware 2.7.26, AND deleted the nodeDB, they started working again. Might sometimes also be a good idea if all fails to apply nrf_erase2.uf2 before firmware update, following the erase-guide on flasher.meshtastic.org (recycle bin symbol) to get a clean filesystem if all else fails.

The only thing that puzzled me: when updating via nrfConnect App Android -ota.zip, something that always works with DFU / number of packets 5 and RAK4631 devices: the Seeed Solar P1 did not reboot after a complete and successful OTA upload of firmware. Manual power down and power up was required.

Might this be a general problem with Seeed / Xiao-type boatloaders? I mean, once I force-started via power down / up, the node Seeed Solar with the new firmware 2.7.26 was available right away. Not good when a node is up high on a mast :-)