feat(linter): enhance no-unused-vars

[camc314]

typescript-eslint updated it's logic to report if a var is only used as a type, we should do the same

https://playground.oxc.rs/#eNo9TzEOgzAM/ErkhYWhUHUJ6tqlX2AJ1EFIIUaOaUGIvzdpFSbfWee78w49aOjJB1GdYXVXuyXSqqjqa3E0rW89rjOxKNlmVM8oSIBsUjdQAoHegRefRti8mBW08IIluNFLxqGnGU+yTR25zISND5Z4Am2NC3iUMBsOyNEx4uSS8Sn9LUAMDxgjAEN9qW4QFT29cMBUJpJp9KMd83H8UZjcw9EnHb+ROwqx0z/0OL69YFut

https://github.com/typescript-eslint/typescript-eslint/pull/9330/files#diff-5e72d53d5a699795e22013142ec416b4e31cedc1881baba6a225faa7bad499c8

[Dunqing]

While working on #8465, I also discovered that this rule does not yet support reporting this error.

I thought it was not hard for oxlint to support this until I found this edge case would be hard because when need to know whether it has a same name with a type alias declaration and it is exported, if it does, then we don't need to report this error. I suspect this is a bug in typescript-eslint, because only type A is exported not const A, before doing this, we need to confirm in typescript-eslint if it is a correct behavior!

It would be appreciated if you could help us to confirm this! @JoshuaKGoldberg

[JoshuaKGoldberg]

Heya! I'm not 100% following what you mean. But if you think it's a bug in typescript-eslint, could you please file an issue on us? 🙂

[Dunqing]

Heya! I'm not 100% following what you mean. But if you think it's a bug in typescript-eslint, could you please file an issue on us? 🙂

Sorry, I did not explain clearly, I will file an issue later! Thanks!