[python] Support schema evolution of nested struct sub-fields

[TheR1sing3un]

Purpose

Follow-up to #8126, which made read-time schema evolution align top-level columns by field id. This extends the same id-based alignment to sub-fields inside a ROW (including a ROW nested in an ARRAY/MAP).

Before this PR, nested sub-field evolution didn't work: adding a sub-field silently created a top-level column, and rename/drop/update-type failed, because only the last name in the path was matched.

Now a dotted path like mv.value is resolved recursively, so for a column mv ROW<version BIGINT, value STRING>:

• add a sub-field → old rows read NULL for it;
• rename a sub-field → data follows the field id, not the name;
• drop a sub-field → its old data is not revived;
• update type of a sub-field → cast at read time.

# mv ROW<version BIGINT, value STRING>
catalog.alter_table("db.t", [SchemaChange.rename_column(["mv", "value"], "val")])
# files written under mv.value are read back as mv.val with the same data

How it works:

• Nested sub-fields get globally-unique field ids at create time; highestFieldId is computed recursively so nested and top-level ids never collide.
• Schema changes (add / rename / drop / update-type / update-nullability / update-comment) recurse along the field-name path, transparently through ARRAY/MAP wrappers.
• update column type is validated against the cast-support rules.
• The read path aligns nested sub-fields by id — reorder, pad missing with NULL, follow renames, cast changed types — recursing into struct / array / map.

Tests

New cases in schema_evolution_nested_read_test.py:

• nested add / rename / drop / update-type read round-trips, on append-only and primary-key tables;
• sub-fields of ARRAY<ROW> and MAP<.,ROW>;
• the nested field-id model (global uniqueness, recursive highestFieldId, duplicate detection);
• the type-cast support rules.