feat: data pipeline syntax

**BEFORE**:
```python
class Dataset:
# Pipeline syntax
data: DataFrame >> normalize >> remove_outliers >> scale
```

**AFTER**:
```python
# After (transpiled.py)
from typing import Callable
from dataclasses import dataclass
from pandas import DataFrame
from vyper.pipeline import Pipeline, Transform

@dataclass
class Dataset:
_data: DataFrame | None = None

@property
def data(self) -> DataFrame:
if self._data is None:
self._data = (Pipeline()
.then(normalize)
.then(remove_outliers)
.then(scale)
.execute(self._raw_data))
return self._data

@data.setter
def data(self, value: DataFrame):
self._raw_data = value
self._data = None # Invalidate cache
```

Start from the issue's BEFORE and AFTER Python examples to understand the requested pipeline syntax and transpiled output. Done means the shown `>>` pipeline form can produce the corresponding `Pipeline().then(...).execute(...)` behavior, including the demonstrated cache invalidation semantics.

Written by the indexing model from the issue text.