Execution panic for valid 64-column GROUPING SETS #26134

Describe the bug

A GROUPING SETS query with 64 distinct grouping columns can panic during execution, even when no grouping set is repeated.

group_id_array packs the duplicate ordinal using ordinal << n. With 64 columns and ordinal zero, this evaluates 0u64 << 64 and panics in a checked build. The layout itself fits UInt64.

To reproduce

Generate the SQL below and run its output in a debug DataFusion CLI:

n = 64
cols = ", ".join(f"c{i} INTEGER" for i in range(n))
keys = ", ".join(f"c{i}" for i in range(n))
row = ", ".join(str(i + 1) for i in range(n))
print(f"CREATE TABLE wide_keys ({cols});")
print(f"INSERT INTO wide_keys VALUES ({row});")
print(f"SELECT COUNT(*) FROM wide_keys "
      f"GROUP BY GROUPING SETS (({keys}), ());")

Expected behavior

The query should return two rows, each with COUNT(*) = 1. Layouts requiring more than 64 total bits should continue to return NotImplemented.

Additional context

The shift overflow is confirmed by focused regression tests on commit 102a1628592121d96bfe3d09d80e63180eef288a.