Make learned positional embedding optional

Summary:
- Add learned positional embedding binary flag to masked LM model.
- Add base arch config for masked LM model which sets all the binary parameters to False. Otherwise some of the binary flag parameters will always be override by config in `xlm_architecture` (e.g. encoder_learned_pos)

Reviewed By: liezl200

Differential Revision: D15054487

fbshipit-source-id: d78827f352b9160a89c9dc4f45b9fce15a2f234d