Fix mishandling of exceptions in UTC

[ArnyminerZ]

Closes https://github.com/bitfireAT/davx5/issues/898

The TemporalAdapter.isBefore call in ICalendarGenerator.write crashed when comparing a UTC Instant (for example 2025-08-22T03:30:00Z from an exception's DTSTART) against a ZonedDateTime (from the main event's DTSTART), because ical4j internally tries ZonedDateTime.from(Instant) which always throws DateTimeException.

Added a failing test that showcases the issue, and a fix for it.

[Copilot]

Pull request overview

Fixes a crash in ICalendarGenerator.write when determining the earliest DTSTART across a main event and its recurrence exceptions, specifically when exceptions use a UTC Instant while the main event uses a ZonedDateTime (DAVx5 #898 scenario).

Changes:

• Add a regression test covering a recurring event whose exception DTSTART is a UTC Instant.
• Normalize temporals before calling TemporalAdapter.isBefore to avoid DateTimeException when comparing mixed temporal types.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
lib/src/main/kotlin/at/bitfire/synctools/icalendar/ICalendarGenerator.kt	Normalizes exception/main DTSTART values prior to comparison to prevent crashes on mixed temporal types.
lib/src/test/kotlin/at/bitfire/synctools/icalendar/ICalendarGeneratorTest.kt	Adds a test reproducing the UTC Instant exception DTSTART crash and asserting calendar output is produced.

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated no new comments.

[rfc2822]

Is there an issue or did you discover that by chance?

Is it related to #424 somehow (that is also about crash with Instant)?

[devvv4ever]

Is there an issue or did you discover that by chance?

Is it related to #424 somehow (that is also about crash with Instant)?

Issue is https://github.com/bitfireAT/davx5/issues/898 (which I forgot to add to Development Plan ... sorry - did so now)

[ArnyminerZ]

Is there an issue or did you discover that by chance?

No, this is a fix for https://github.com/bitfireAT/davx5/issues/898

Is it related to #424 somehow (that is also about crash with Instant)?

I don't think so, they come from different places

[rfc2822]

No, this is a fix for bitfireAT/davx5#898

Ah, nice, wasn't aware of that one! I linked it to the PR.

[ArnyminerZ]

I linked it to the PR.

I didn't link it originally so that we remember to update the synctools version in DAVx⁵ as well, but I guess it will eventually get updated anyway so