*   Fix `ANEURALNETWORKS_BAD_DATA` execution failures with
        `sum`/`max`/`min`/`reduce` operations with `scalar` inputs.
*   Hexagon
    *   TFLite Hexagon Delegate out of experimental.
    *   Experimental `int8` support for most hexagon ops.
    *   Experimental per-channel quant support for `conv` in Hexagon delegate.
    *   Support dynamic batch size in C++ API.
*   CoreML
    *   Opensource CoreML delegate
*   Misc