fix(preprocessor): raise a clear error on duplicate column names

[ChrisW09]

Fixes #37.

Problem

dup = pd.DataFrame(np.column_stack([rng.normal(size=50)] * 2), columns=["a", "a"])
Preprocessor(numerical_method="minmax").fit(dup, rng.normal(size=50))

AttributeError: 'DataFrame' object has no attribute 'dtype'
  ... pretab/preprocessor.py, line 307, in _detect_column_types

X[col] returns a DataFrame rather than a Series when col is duplicated, so the dtype
inspection has nothing to read. (nunique() on the preceding line happens to work on a
frame, which is why the failure surfaces one line later and looks unrelated.)

Fix

Detect duplicates up front and raise a typed error naming them:

PretabDataError: Duplicate column names are not supported: ['a'].
Fix: rename the columns so every name is unique.

Rejecting rather than accommodating is the honest outcome here: the ColumnTransformer this
builds keys its transformers by column name (f"num_{feature}"), so duplicate labels could
not be routed unambiguously even if type detection coped with them.

Tests

Three added to tests/test_exceptions.py: the reported single-duplicate case, a frame with
two distinct duplicated labels (asserting both are listed), and a unique-column frame to
confirm the check is not over-eager.

Full suite: 483 passed, 9 xfailed. ruff check clean on changed files; pyright unchanged at
its pre-existing 71 errors.

🤖 Generated with Claude Code