converter: Improve bit_position computation

converter: Improve bit_position computation

- Fix wrong bit_position swap computed:
Need to multiply then lenght in bytes by 8 to get it for bits.

- Add a continental bit_position function:
This allow to get the correct bit_position using the weird Continental.
bit numbering method where the Frame is read using little endianness
and bit count using a big endianness

- Use define CHAR_BIT instead of hardcoded 8 value

Change-Id: I8d458ab457f80e5fa116af942acfceaaef51b763
Signed-off-by: Romain Forlot <romain.forlot@iot.bzh>