allocator: implement output buffering to reduce frequent log.VEvent calls

**Is your feature request related to a problem? Please describe.**

When investigating delays in an allocator simulator test, we notice that a
significant amount of time were spent on `log.VEventf`. The issue originates
from `rankedCandidateListForRebalancing ` repeatedly calling
`IsStoreReadyForRoutineReplicaTransfer ` on all stores that pass
`StoreFilterThrottled `. As a short term mitigation, we removed the
`log.VEventf` for less interesting cases where store are alive. However, this
could indicate a prevalent issue.

A longer term strategy is to delve into different sections of the allocator code
and implement output buffering at a higher level to reduce frequency of
`log.VEventf`calls.

Jira issue: CRDB-30033

Epic CRDB-54646