[fix](test) fix time series compaction rowset count

[shuke987]

Summary

Fix compaction/test_time_series_compaction_level2.groovy rowset count assertions for branch-4.1 environments that force table replication to 3.

The test counts rowsets across every tablet replica returned by SHOW TABLETS. In the failing x64 branch-4.1 NonConcurrent build, force_olap_table_replication_num=3 overrides the case's replication_num=1, so the base rowset exists on each replica too. The previous assertions multiplied inserted/compacted rowsets by replicaNum but counted the base rowset only once, causing the first check to expect 31 while the actual rowset count is 33.

This patch applies replicaNum consistently to:

• the per-round pre-compaction rowset count
• the post-loop level-1 rowset count
• the final base + level-2 rowset count

Testing

• git diff --check
• Checked TeamCity occurrence: build 201293, occurrence 2000000344
• Runtime regression validation: pending run buildall

[hello-stephen]

Thank you for your contribution to Apache Doris.
Don't know what should be done next? See How to process your PR.

Please clearly describe your PR:

1. What problem was fixed (it's best to include specific error reporting information). How it was fixed.
2. Which behaviors were modified. What was the previous behavior, what is it now, why was it modified, and what possible impacts might there be.
3. What features were added. Why was this function added?
4. Which code was refactored and why was this part of the code refactored?
5. Which functions were optimized and what is the difference before and after the optimization?

[shuke987]

run buildall

[shuke987]

run buildall