Support dumping SQL files to a local filesystem

[kennytm]

Create an MVP for this project, showing performance and correctness. See pingcap/community#123 for the full list of expectations.

Checklist:

• Make sure your architecture takes account of all features in Phases 1 and 2. Implementing those features in the future should not require a complete refactoring.
• Resulting data are sorted by primary key (_tidb_rowid, if exists, is always considered as the primary key.)
• SQL files are split into size close to the given configuration
• Single tables can be dumped in parallel, if a primary key or unique btree key exists (Postponed into Support parallel dumping of a single table #12, and TiDB may not be able to handle it)
• Consistency: dumping a snapshot instead of live data (either acquire a read lock or ignore new updates)
• Error handling and recovery
• Support all data types allowed by TiDB (text, blob, int, float, decimal, bit, enum, set, timestamp, datetime, date, time, year, json)
• Does not fail with edge cases for data (e.g. naughty strings, character sets, time zones)
• Properly handle special features (e.g. special characters in table name, table partitions, views, generated columns, MariaDB sequences, etc.)
• Has adequate logging and Prometheus metrics
• Unit test coverage ≥ 75%

Score

• 6200 5000