[tidb-x] add index requires an advance estimate of disk capacity usage

## Bug Report

Please answer these questions before submitting your issue. Thanks!

### 1. Minimal reproduce step (Required)
1，create cloud cluster with 3 * tikv-1.4TB
2，add index for one big table (800GB/TiKV)
3，add index failed with error
[2026/03/31 01:37:20.984 +00:00] [WARN] [index.go:1668] ["run add index job failed, convert job to rollback"] [keyspaceName=10053164508501693964] [category=ddl] [job="ID:34, Type:add index, State:running, SchemaState:write reorganization, SchemaID:21, TableID:32, RowCount:5000000000, ArgLen:1, start time: 2026-03-30 23:57:47.061 +0000 UTC, Err:, ErrCount:0, SnapshotVersion:465283335392067694, Version: v2, UniqueWarnings:0, Multi-Schema Change:true, Revertible:true"] [error="[Ingest:StoreDiskFull][Ingest:StoreDiskFull](ingest SST ID 465284884663107690)"]

### 2. What did you expect to see? (Required)
add index can advance estimate of disk capacity usage

### 3. What did you see instead (Required)
add index failed

### 4. What is your TiDB version? (Required)