Fix early exit of /log/public/historical/range when there are empty pages

[cjen1-msft]

When we reduced the logging page size in a recent test we had a failure where the range query did not turn up the entry.

This is because if there is is an empty page (not the first page), then the range query would exit early rather than directing the client to continue iterating past it.

One possible optimisation here is that the @nextlink could skip across that empty page, but this version is obviously correct, if slightly inefficient.

This was tested using a reduced page size of 10, and submitting 10 txs and their signatures. with an 11th tx which we then queried for.

[Copilot]

Pull request overview

Fixes the logging sample app’s historical range pagination so clients continue iterating across empty pages, rather than terminating early when a non-first page contains no entries.

Changes:

• Always emit @nextLink (via response.next_link) when the requested historical range isn’t fully covered yet, even if the next page is empty.
• Keep prefetch behaviour only when the next page is known (via the index) to contain interesting seqnos.
• Add a changelog note describing the fix.

Custom instructions used:

• .github/copilot-instructions.md
• .github/instructions/reviewing.instructions.md
• .github/instructions/changelog.instructions.md

Reviewed changes

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

File	Description
samples/apps/logging/logging.cpp	Removes early-exit pagination optimisation by always returning a next_link when more of the requested seqno range remains.
CHANGELOG.md	Adds a “Fixed” entry describing the logging app pagination fix.

[Copilot]

Pull request overview

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

[cjen1-msft]

Waiting for the mypy fix in #7867