Update ADBC Client Docs

[henryharbeck]

After #81, it seems the Python ADBC driver is packaged with DuckDB for Python. Can the docs be updated to reflect this. Specifically that the "Setting Up the DuckDB ADBC Driver" no longer applies for Python.

FWIW, thanks to the Arrow PyCapsule interface, PyArrow is no longer required for full functionality with ADBC either. For example, reading and writing Arrow data no longer requires PyArrow. Converting data between Python and Arrow either requires PyArrow (full functionality) or Polars as a recently added backend (partial functionality).

Here's an example of this reading and writing Arrow data with only duckdb and adbc-driver-manager installed

# /// script
# requires-python = ">=3.11"
# dependencies = [
#     "adbc-driver-manager",
#     "duckdb",
# ]
# ///
import importlib.util

if importlib.util.find_spec("pyarrow") is not None:
    raise RuntimeError("Uninstall PyArrow!")

import adbc_driver_duckdb.dbapi
import duckdb

with adbc_driver_duckdb.dbapi.connect("my.db") as conn, conn.cursor() as cur:
    cur.execute("SELECT 42 AS a")
    # Fetch the result as an object implementing the Arrow PyCapsule interface.
    arrow = cur.fetch_arrow()
    rel = duckdb.from_arrow(arrow)
    print(rel)
    # ┌───────┐
    # │   a   │
    # │ int32 │
    # ├───────┤
    # │    42 │
    # └───────┘

    other_rel = duckdb.sql("SELECT 'foo' AS b")
    cur.execute("DROP TABLE IF EXISTS my_new_table")
    cur.adbc_ingest("my_new_table", other_rel)
    conn.commit()

# Prove we persisted the data
new_conn = duckdb.connect("my.db")
print(new_conn.sql("FROM my_new_table"))
# ┌─────────┐
# │    b    │
# │ varchar │
# ├─────────┤
# │ foo     │
# └─────────┘