`TextVectorization`. Multi-hot encoding will no longer automatically
            uprank rank 1 inputs, so these layers can now multi-hot encode
            unbatched multi-dimensional samples.
        *   Added a new output mode `"one_hot"` for `CategoryEncoding`,
            `StringLookup`, `IntegerLookup`, which will encode each element in

$0, Y1H // VZERO Y1H VSPLTISB $0, Y1L // VZERO Y1L loop_select: LXVD2X (P1ptr+R0), X2H LXVD2X (P1ptr+R16), X2L LXVD2X (P1ptr+R17), Y2H LXVD2X (P1ptr+R18), Y2L VCMPEQUD SEL2, IDX, SEL1 // Compare against idx VSEL X1L, X2L, SEL1, X1L // Select if idx matched VSEL X1H, X2H, SEL1, X1H VSEL Y1L, Y2L, SEL1, Y1L VSEL Y1H, Y2H, SEL1, Y1H VADDUBM SEL2, ONE, SEL2 // Increment SEL2 bytes by 1 ADD $64, P1ptr // Next chunk BDNZ loop_select STXVD2X X1H, (P3ptr+R0) STXVD2X X1L, (P3ptr+R16) STXVD2X Y1H, (P3ptr+R17)...