Support hexadecimal Default Appearance

[btecu]

getDefaultAppearance is used to determine font size and the style among other things.
Some pdf files can have the DA property as a hexadecimal, causing the library to have a un-parsable DA value.

Bonus: Make return value match the function signature (string | undefined).

[Hopding]

@btecu Do you have an example of a PDF that contains hex string DAs? Also, I'm all for making this change, but I would like to create some unit tests around this method to ensure the functionality isn't broken or removed down the line.

[btecu]

@Hopding I sent you an example document.

[Hopding]

Thanks @btecu! Would you please create a unit test for each of the altered methods? For example, tests/core/acroform/PDFAcroField.spec.ts would load the document you shared with me (thus it would need to be added to the assets/pdfs directory), read the raw appearance strings and assert that they are hex strings, and then call the getDefaultAppearance() methods and assert that the correct value is returned. Note that we'll need to do this for both PDFAcroField and PDFWidgetAnnotation instances.

[btecu]

@Hopding I cannot add the PDF that I shared with you since is not a public document, and I don't know how to add a test for it without the PDF.

[Hopding]

@btecu I added tests in #693. These changes will go out in the next release of pdf-lib. Thanks for your contribution! 🎉

[Hopding]

The changes in this PR were just released in v1.13.0.