parse: Add ParseError and ParseErrorException for error reporting

parse: Ignore whitespace in text, instead join words with spaces ...

This cuts down on complexity as now only the tokenizer deals with
whitespace and makes whitespace no longer a factor in the language.

tokenize: Move syntax to its own module

parse: Add read_token function

parse: Move Syntax to tokenize and replace Token ...

These are basically the same classes so merge them

parse: Don't nest tokens in values ...

Now that we don't need to copy the TokenType we can just copy the value
directly.

tokenize: Remove TokenType enum ...

In practice this wasn't used: The parse code always checks the value AND
the type. So just check the value only.

tests: parse: Draw text syntax tokens ...

This requires fixing a bug for comparing values of different types.

parse: tests: Clarify errors and tests

parse: Check token types when parsing text

parse: Add parse_text to parse a text syntax node ...

No error handling tests are added yet.

parse: Rename TokenStream to SyntaxStream

parse: Require the caller to copy the TokenStream list ...

This is more Pythonic I think.

parse: Add TokenStream class ...

This will be used for stateful parsing, with lookahead in future.

parse: Don't strip non-token syntax ...

This bug was picked up by mutmut.

parse: Use syntax for strip_whitespace

parse: Add import_tokens to import tokens to syntax

parse: Add Syntax node ...

Currently this just holds a token, but will eventually hold parsed
syntax tree values.

parse: Rename remove_whitespace to strip_whitespace

parse: Add parse wrapper function

parse: Add remove_whitespace pass

parse: Move to oldparse ...

This is in preparation of writing a new parser.
Once the new parser is ready then this code can be junked.

Change license to LGPL-2.1-only

src: parse: Fix uppercase repr value

parse: Use __str__ for ParseError printing ...

Exceptions don't use __repr__ but __str__ for printing

tests: lexer: Check error location

src: parse: Don't assign unused variables ...

Found by flake8.

parse: Strip whitespace in text tokens

parse: Replace 'Begin' with 'Start' in keywords ...

Okay, it's to the point I'm writing bugs in tests due to this.
Just change it already.

parse: Read text literal up to beginning of EndText ...

Using the end of the last word before EndText misses any trailing
whitespace, and it's a lost cause trying to figure out the offset
from after EndText given it might be whitespace or an end of life.

Instead use an offset from the start of EndText.