drainer/schema: replace drainer's schema with ticdc's schema storage

[lichunzhu]

Feature Request

Is your feature request related to a problem? Please describe:

Drainer will fetch all history DDL jobs and handle almost all of them when drainer starts. However, it will consume a lot of memory and cost a lot of time.

Describe the feature you'd like:

Use ticdc's schema storage package to get table info to avoid loading too many ddl jobs when drainer restarts.

Describe alternatives you've considered:

Teachability, Documentation, Adoption, Migration Strategy:

[lichunzhu]

fix on release-3.0 #1178