Fixes #998 and surpasses #999.

(Can either be for Beta3 or Beta4, whatever suits us best)

We again have three parts, but slightly modified from #999:

String-fallback for Char in convert

We get String-fallback for Char columns in df.convert().to<>()... and col.convertTo<X>(). This means you can now do things like:

enum class EnumClass { A, B }

columnOf('A', 'B').convertTo<EnumClass>()
// previously you could only do this
columnOf('A', 'B').convertToString().convertTo<EnumClass>()

Note: We do keep the old convert behavior of Char -> Int by ASCII code. This is in line with the JVM world. We have parsing now if you want the Char '1' to turn into the Int 1.

Converting from String columns can only result in enum instances (like above), value class instances, or call parse(), so we follow the same behavior for Char columns now.

Char parsing

df.parse() now not only parses string columns, but also char ones (this can be changed if you don't agree, of course).
We also gain:

columnOf('1', '2', '3').parse() // results in DataColumn<Int>
columnOf('T', 'F', 'T').parse() // results in DataColumn<Boolean>

Here I deviate from #999, as

columnOf('a', 'b', 'c').parse() // throws IllegalStateException

// just like
columnOf("aa", "bb", "cc").parse() // throws IllegalStateException

because the parsing didn't do anything. However similar to String columns, there's tryParse() which can return the same type as the input.

convertTo<> { parser {} } Char support

Fixes #998 by trying out all provided String converters if Chars are encountered without a converter.

df.convertTo<Schema>() {
    // can now be used for both String and chars
    parser { /* it: String -> */ MyCustomClass(it) }

    // will be used instead of parser {} if present
    // convert<Char>().with { /* it: Char -> */ ... }
}

I also updated the docs and expanded upon convertTo a lot to prevent confusion, like what was experienced in #998.