Add support for formatting escript files

[sile]

Closes #101

Copilot generated summary

This pull request introduces significant changes to the tokenizer implementation in the efmt_core library, replacing the use of erl_tokenize::Tokenizer with a custom Tokenizer that handles shebang lines. Additionally, support for .escript files has been added. Here are the most important changes:

Tokenizer Implementation:

• efmt_core/src/lib.rs: Replaced erl_tokenize::Tokenizer with the new Tokenizer in the format_text function.
• efmt_core/src/parse.rs: Added TokenOrShebang and Tokenizer modules and made them publicly available.
• efmt_core/src/parse/token_stream.rs: Updated TokenStream to use the new Tokenizer and handle TokenOrShebang variants. [1] [2] [3] [4]
• efmt_core/src/parse/tokenizer.rs: Implemented the new Tokenizer struct and TokenOrShebang enum to handle shebang lines.

Support for .escript Files:

• src/files.rs: Added .escript to the list of format targets and updated the tokenizer usage. [1] [2] [3]
• src/lib.rs: Updated the format_file and format_text functions to use the new Tokenizer. [1] [2]
• tests/lib.rs: Added .escript to the test file extensions.
• tests/testdata/factorial.escript: Added a test .escript file for factorial calculation.