`RandomTranslation`, `RandomRotation`, `RandomHeight`, `RandomWidth`,
    `RandomZoom`, `RandomContrast`
*   Improved **`TextVectorization`** layer, which handles string tokenization,
    n-gram generation, and token encoding
    *   The `TextVectorization` layer now accounts for the mask_token as part of
        the vocabulary size when output_mode='int'. This means that, if you have

0FC6          ; valid                                  # 3.0  TIBETAN SYMBOL PADMA GDAN
0FC7..0FCC    ; valid                  ;      ; NV8    # 3.0  TIBETAN SYMBOL RDO RJE RGYA GRAM..TIBETAN SYMBOL NOR BU BZHI -KHYIL
0FCD          ; disallowed                             # NA   <reserved-0FCD>
0FCE          ; valid                  ;      ; NV8    # 5.1  TIBETAN SIGN RDEL NAG RDEL DKAR