Implement operand for scipy.special.fresnel_zeros

[RandomY-2]

Is your feature request related to a problem? Please describe.
We currently support all scipy special error functions and fresnel integrals except fresnel_zeros, so it would be good to support it as well.

Describe the solution you'd like
As fresnel_zeros only accepts scalar inputs, it is not a typical scipy.special function. Thus it should be treated as something like mt.ones which accepts scalar dimensions and outputs tensors with multiple chunks other than simple arithmetic operands. We would want to implement a separate operand class just for fresnel_zeros, and the operand should produce multiple chunks given options.chunk_store_limit or some chunk_size value provided by the caller.

Additional context

• Scipy special error functions and fresnel integrals: https://docs.scipy.org/doc/scipy/reference/special.html#error-function-and-fresnel-integrals
• Context: Enable TensorTupleOp to support functions with scalar inputs #3157

[dlee992]

BTW, recently I saw JAX and numba-scipy both contain some scipy.special module code for faster execution speed. Does Mars have interface to choose these execution engine? Or have any schedule about this kind of optimization?