fix fvar axis/instance name parsing and writing (fix #687)

[Connum]

Description

Fixes parsing of name IDs in the fvar table (fix #687) and adds parsing/optional writing of the postScriptNameID field.
I also refactored name handling a bit. Having the fvar make function modify the name table by creating new entries (without checking if a name entry already existed) was a bad design choice and shouldn't be done during writing anyway, so I removed the addName() function completely. New names should be added to the name tables and then be referenced by ID in the fvar table. When we add "comfort" functions to the variable features later, these should handle this accordingly, but for now this has to be done manually.

Motivation and Context

Parsing the names for axes/instances in the fvar table did not work correctly in all cases. Name ID values 2 and 17 for the instance name, as well as 6 for the postScriptNameID were not parsed correctly (and postScriptNameID was so far not parsed at all). These are allowed according to the spec:

• https://learn.microsoft.com/en-us/typography/opentype/spec/fvar#:~:text=Values%20of%202%20or%2017%20can%20be%20used
• https://learn.microsoft.com/en-us/typography/opentype/spec/fvar#:~:text=Values%20of%206%20and%200xFFFF%20can%20be%20used

How Has This Been Tested?

Test font provided in #687 and additional tests created.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist:

• I did npm run test and all tests passed green (including code styling checks).
• I have added tests to cover my changes.
• My change requires a change to the documentation.
• I have updated the README accordingly.
• I have read the CONTRIBUTING document.

[eweracs]

Thanks a lot, Constantin! This is a rather crucial fix for a serious bug, is there some way this could be reviewed in the near future?

[Connum]

is there some way this could be reviewed in the near future?

I hope so! Pinging @yne @ILOVEPIE

[yne]

thanks @Connum 💪

[Connum]

Rebased onto current master, needs another review to be able to merge

[Connum]

Oh boy, the rebase broke something... I'll have a look at it later.

[Connum]

@yne ready for final review