[Python] support latin1/utf16 encoding for python string

[chaokunyang]

Feature Request

Fury java serialize string with three encodings:

• latin1: used when all chars are latin1. This will be just a memory copy in jdk11+
• utf16: used when 50%+ chars are not ascii chars
• utf8: used when 50%+ chars are ascii chars

Fury java also use superword and bitmask for 8 bytes ascii check/writing at once, which will make encoding faster.

Here is the fury benchmark result with jdk/kryo/flink string serializer:

For pyfury, we should do similar things, and since pyfury can invoke c++ with low cost, we could implement string encodings using SIMD in c++ and let pyfury wrap that by cython.

Is your feature request related to a problem? Please describe

No response

Describe the solution you'd like

No response

Describe alternatives you've considered

No response

Additional context

#1732
#1754
#1890
#1964

[chaokunyang]

Hi @penguin-wwy , are you interested in this issue?

[penguin-wwy]

Hi @penguin-wwy , are you interested in this issue?

Okay, I will implement it.