change String::to_bytes() to String::encode()

[bobzhang]

The current String::to_bytes is misleading and confusing to AI given that MoonBit is utf16 encoded, we are gonna use x.encode(Utf8|Utf16le|Utf16be|Utfl16) instead 

I saw lots of code generated by AI which use .to_bytes in a wrong way
https://chatgpt.com/share/68b011a2-0308-8000-9544-10b586c22313

[hackwaly]

How about:

str.to_bytes(encoding=Utf8) // make `encoding` a non-optional named parameter

if changes .to_bytes to .encode, I suspect it should be in @strconv package.

[peter-jerry-ye]

fn Buffer::write_char_utf8
fn Buffer::write_char_utf16le
fn Buffer::write_string_utf8
fn Buffer::write_string_utf16le

[hackwaly]

fn Buffer::write_char_utf8
fn Buffer::write_char_utf16le
fn Buffer::write_string_utf8
fn Buffer::write_string_utf16le

So, we need to deprecate String::to_bytes?