[server] Optimize RemoteLogFetcher with async prefetch for recovery

[Kaixuan-Duan]

Purpose

Linked issue: close #3091
This PR improves KV recovery performance by reducing wait time between remote log segments in RemoteLogFetcher.

Brief change log

• Add a dedicated single-thread download executor for async prefetch.
• Prefetch the next fetchable remote segment while consuming the current one.
• Reuse prefetched file when available; fallback to synchronous download on prefetch failure.
• Ensure cleanup on close: cancel in-flight prefetch, close active iterator/resources, shutdown executor.
• Add a regression test for repeated fetch() to ensure previous iterator cleanup.

Tests

• ./mvnw -pl fluss-server -Dtest=RemoteLogFetcherTest -DfailIfNoTests=false -Dspotless.check.skip=true test

API and Format
No API change. No storage/log format change.

Documentation
No user-facing feature. No documentation update required.