Implement FT.CURSOR command (FT.SEARCH, FT.AGGREGATE)

[vyavdoshenko]

Currently, Dragonfly does not support the FT.CURSOR command for incremental retrieval of large result sets.
This feature is essential for handling large aggregation or search results efficiently without loading the entire dataset into memory at once.

Expected behavior:
Implement the following subcommands:

• FT.CURSOR READ <cursor_id> [COUNT ]
  Returns the next batch of results for an existing cursor.
• FT.CURSOR DEL <cursor_id>
  Explicitly deletes a cursor and frees server-side resources.

Support for creating cursors should be added to:

• FT.AGGREGATE with WITHCURSOR [COUNT ]
• FT.SEARCH with WITHCURSOR [COUNT ].

Why it matters:

• Enables incremental result fetching for large queries.
• Improves client performance and memory usage.