Lisp lexer (scanner) implemented in GO used by jig/lisp and jig/tinylisp

Code adapted from the Go text/scanner standard package, and mostly compatible with it (beside the fact this is for Lisp and that is for Go syntax).

Major differences:

• identifier allowed characters is more expansive than Go is (e.g. +, -, *host-name*, or <= are valid identifiers)
• negative integers or negative floats are supported as a single token (of type Int or Float respectively)
• character tokens are not supported anymore (e.g. 'A') as ' is used as synonym of quote
• parsing errors are not printed to os.Stdout by default
• raw strings in jig/lisp are quoted with ¬ character (instead of `) as ` is used as synonym of quasiquote. Raw strings might include ¬ by doublind them ¬¬
• #{ is specially handled to support Lisp set literals
• ~, @ and ~@ are specially handled to support them as synonyms of unquote, deref and splice-unquote respectively
• support of Lisp keywords (e.g. :key) as a single token of type Keyword

See pkg.go.dev/jig/scanner