[sinks] Set with_snapshot=false even during history reduction

[bkirwi]

Motivation

In #31152, we added an optimization for with_snapshot that kicks in when environmentd restarts, such as during a deploy. However, if only clusterd restarts, we will use the sink definition cached in the history implementation... and if the first run of the sink needed the snapshot, then we'll fetch the snapshot after a restart too.

This PR also applies the optimization in the history, so the snapshot can be skipped when even the first run of a sink is restarted. I feel less confident in this change, since the "chain of reasoning" that implies it is safe is a bit more complex. But it may be worthwhile to avoid the operational surprise.

Happy for feedback on this point however!

Checklist

• This PR has adequate test coverage / QA involvement has been duly considered. (trigger-ci for additional test/nightly runs)
• This PR has an associated up-to-date design doc, is a design doc (template), or is sufficiently small to not require a design.
• If this PR evolves an existing $T ⇔ Proto$T mapping (possibly in a backwards-incompatible way), then it is tagged with a T-proto label.
• If this PR will require changes to cloud orchestration or tests, there is a companion cloud PR to account for those changes that is tagged with the release-blocker label (example).
• If this PR includes major user-facing behavior changes, I have pinged the relevant PM to schedule a changelog post.

[bkirwi]

Nightlies - the first run was pretty red, but it seems to be unrelated breakage in main... rerunning the failed tests after a rebase seems greener.