Clarify `R_TIMESTAMP_OFFSET` behavior

[bruno-f-cruz]

When evaluating #23, several questions regarding register R_TIMESTAMP_OFFSET (15) came up.
@filcarv would you mind clarifying:

1. How is this register implemented? The value is a U8, but the increments are 500us according to the spec.
2. Can this register be used to update values with a negative offset?
3. How many boards currently implement this feature?

We may want to quickly discuss this during a future SRM, but fo rnow it would be great to have some background information.

[filcarv]

1. It's the expected latency between the computer and the device. When the timestamp is written, the device synchronizes itself with the timestamp + this offset (because it was sent some time ago).
2. I didn't understand the question, sorry.
3. The RfidReader I believe it's the only device using it.

[bruno-f-cruz]

During https://github.com/orgs/harp-tech/discussions/46 discussed:
The register's value work by converting each unitary value of the payload to 500us increments (0x01 -> 500us, 0x02 -> 1ms, etc...)
This register is not used at all during hw clock sync

Only the RFID currently uses this register. As a result the proposal is to move this to an application register of this device and deprecate the register in the core.

[bruno-f-cruz]

Superseded by #113