N85 is a binary-to-ASCII encoding based on Ascii85 but more suitable for use as strings
(i.e. excludes \, ' and ") and with a simpler implementation than the other variants.
Every 4 bytes is mapped to 5 characters, adding ~25% of storage overhead. For comparison, base64 maps every 3 bytes to 4 characters, adding ~33% overhead.
```rust let s = "Hello, world!";
let enc = n85::encodestring(s.asref()); assert_eq!(&enc, "Yb(qJ[NH@N0AO?HI(");
let dec = n85::decodevec(enc.asref()).unwrap(); asserteq!(&dec, s.asbytes()); ```
An arbitrary byte string is split into chunks for 32-bit little endian integers. The last chunk is padded with zeroes.
Every integer is 5 times divided by 85, giving 5 remainders.
40 (() is added to each remainder.
If the result is equal or greater to 92 (\), 1 is added.
If the last chunk is 1, 2, 3 or 4 byte large, 2, 3, 4 or 5 characters are used for encoding respectively.