M24C16 EEPROM has issues with addressing and block size

[hairymnstr]

One of the recommended EEPROMS (M24C16) has issues:

1. The current method for identifying an 8 bit addressed EEPROM is to write a 16 bit address then try a read and see if it fails. This writes a byte (16 bit address is interpreted as 8 bit address and 8 bit data) to the EEPROM potentially corrupting it.
2. The default block size for the littlefs partition is 512 bytes, on the M24C16 this means you only get 3 blocks after the header which isn't enough to make a decent sized file.
3. The EEPROM appears as 8 separate 256 byte (one byte addressable) EEPROMs which upsets some layers of the partitioning code making it think the EEPROM is too small to format.
4. The multi-address thing also appears to raise issues for other I²C devices in the 0x5x address range.

I'm gathering all these into this issue and have a branch to work on it here: https://github.com/emfcamp/badge-2024-software/tree/eeprom_stuff

That branch makes the M24C16 work okay for me, but I need to test it against other EEPROMs to check I've not broken them, and I don't think it addresses point 4 yet.

[hairymnstr]

Fixes written and tested see #82

[hairymnstr]

Fixed in 1.8.0