Fix apdu construction and chunking in MP::writeMemory (#152)

[tensoid]

Changes

• Removed the first 14 bytes (code, flags, dataBlockStartAddr, startAddress, endAddress) from the payload construction in MP::writeMemory since these fields are already handled by the MC::writeMemory.

• Removed setAddress method as it is no longer used in the updated implementation.

• Improved chunking logic in the memory write operation:
  Instead of always using the maximum APDU length, the range is now capped based on the remaining data and the actual max APDU size.

Anything else you would like me to change?
Also a quick question, is the max apdu size written down somewhere in the spec, or where did you get that number from? I cant seem to find it in the spec.

Fixes #152

[bmalinowsky]

Thanks for addressing this issue.

I added a few commits to make memory services more deterministic wrt which service is used and what apdu size should be supported. (Note, Calimero currently doesn't check line couplers for configuration. And the check most likely introduces some delay.)

The max. apdu size is device & network specific. All devices have to support an apdu of 15 bytes, but new services are written for 254 (the max. that the device property might return).