[pyupgrade] Do not report when a UTF-8 comment is followed by a non-UTF-8 one (UP009)

[InSyncWithFoo]

Summary

Resolves #14704.

Test Plan

cargo nextest run and cargo insta test.

[github-actions]

ruff-ecosystem results

Linter (stable)

✅ ecosystem check detected no linter changes.

Linter (preview)

✅ ecosystem check detected no linter changes.

Formatter (stable)

✅ ecosystem check detected no format changes.

Formatter (preview)

✅ ecosystem check detected no format changes.

[MichaReiser]

We should address the case when there are multiple coding comments (that might also be on lines other than the first two)

[MichaReiser]

Wow, thanks for implementing the count_lines method!

There's one more edge case that needs handling

#! /usr/bin/env/python
# -*- coding: utf-8 -*-
# -*- coding: ascii -*-

The second comment gets flagged as unnecessary, which is incorrect because removing it would change the encoding to ascii

[InSyncWithFoo]

Cases to handle, as a table:

1	2	3	E
-	-	-
-	-	UTF-8
-	-	Other
-	UTF-8	-	x
-	UTF-8	UTF-8	x
-	UTF-8	Other
-	Other	-
-	Other	UTF-8
-	Other	Other
UTF-8	-	-	x
UTF-8	-	UTF-8	x
UTF-8	-	Other	x
UTF-8	UTF-8	-	x
UTF-8	UTF-8	UTF-8	x
UTF-8	UTF-8	Other	x
UTF-8	Other	-
UTF-8	Other	UTF-8
UTF-8	Other	Other
Other	-	-
Other	-	UTF-8
Other	-	Other
Other	UTF-8	-
Other	UTF-8	UTF-8
Other	UTF-8	Other
Other	Other	-
Other	Other	UTF-8
Other	Other	Other

[MichaReiser]

@InSyncWithFoo what's the status of this PR? Is it ready for re-review?

[InSyncWithFoo]

@MichaReiser It is. Please do.

[MichaReiser]

Thanks for addressing the feedback.

The rule currently incorrectly flags the following comments:

x = 10
# -*- coding: utf-8 -*-
# -*- coding: utf-8 -*-

x = 10
# -*- coding: utf-8 -*-

I think we need to make the comment_ranges iterator more stateful. It should never return Some after it has seen any non-comment. It might be worth implementing a custom iterator and tracking some internal state.