Fix wrong merged table check in JFRDataQueryEsDAO

[currenjin]

What's this PR about?

JFRDataQueryEsDAO.getByTaskIdAndInstancesAndEvent() passes the wrong index name to isMergedTable().

// before
if (IndexController.LogicIndicesRegister.isMergedTable(AsyncProfilerTaskRecord.INDEX_NAME)) {
// after
if (IndexController.LogicIndicesRegister.isMergedTable(JFRProfilingDataRecord.INDEX_NAME)) {

The method queries jfr_profiling_data, but the merged table check uses async_profiler_task. Every other ES DAO passes its own INDEX_NAME. This one was copied from AsyncProfilerTaskQueryEsDAO without updating.

With the default config (logicSharding=false), both tables resolve to records-all, so isMergedTable() returns true for both and the behavior happens to be correct. But with logicSharding=true, the two tables can have different physical names, and the check would produce the wrong result.

Which issue does this PR relate to?

N/A - found by code review.

[Copilot]

Pull request overview

Fixes an Elasticsearch DAO query guard so JFRDataQueryEsDAO.getByTaskIdAndInstancesAndEvent() correctly checks whether the JFR profiling data index is configured as a merged table, ensuring the merged-table discriminator filter is applied (or skipped) against the right logical index.

Changes:

• Switch isMergedTable() to use JFRProfilingDataRecord.INDEX_NAME instead of AsyncProfilerTaskRecord.INDEX_NAME.
• Remove the now-unused AsyncProfilerTaskRecord import.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.