Fix: Convert to numpy array before downsampling

[jayceslesar]

Was attempting to use this with non-numpy arrays in a project https://github.com/jayceslesar/hyperplot and was having a small inconvenience where needed to convert to numpy arrays as that is an underlying assumption of this library - this should make that assumption more robust. Would also be interesting to integrate this into polars using the UDf/Pipe methodology but might want a separation of concerns there as this library does what it needs to really well and polars does what it needs to really well and ideally all the magic happens on the rust side of things anyways

Essentially this would fail:

from datetime import datetime

import numpy as np
import polars as pl

from tsdownsample import LTTBDownsampler

data = {
    "timestamp": [
        datetime(year=2023, month=1, day=1),
        datetime(year=2023, month=1, day=2),
        datetime(year=2023, month=1, day=3),
    ],
    "value": [1, 2, 3],
}
df = pl.DataFrame(data)

indices = LTTBDownsampler().downsample(df["timestamp"], df["value"], n_out=3)

with

python3 testing.py

~/projects/tsdownsample on 🌱 feature/polars-support [$?] is 📦 v0.1.0 via 🐍 v3.10.6 (venv) via 🦀 v1.69.0-nightly on ☁️  jslesarbeta.team(beta-data-warehouse-test) 
❯ python3 testing.py
Traceback (most recent call last):
  File "/Users/jslesar/projects/tsdownsample/testing.py", line 18, in <module>
    indices = LTTBDownsampler().downsample(df["timestamp"], df["value"], n_out=3)
  File "/Users/jslesar/projects/tsdownsample/tsdownsample/downsampling_interface.py", line 322, in downsample
    return super().downsample(*args, n_out=n_out, parallel=parallel, **kwargs)
  File "/Users/jslesar/projects/tsdownsample/tsdownsample/downsampling_interface.py", line 107, in downsample
    x, y = self._check_valid_downsample_args(*args)
  File "/Users/jslesar/projects/tsdownsample/tsdownsample/downsampling_interface.py", line 54, in _check_valid_downsample_args
    if y.ndim != 1:
  File "/Users/jslesar/projects/tsdownsample/venv/lib/python3.10/site-packages/polars/utils/decorators.py", line 169, in _redirecting_getattr_
    return obj.__getattribute__(item)
AttributeError: 'Series' object has no attribute 'ndim'

And changing the line to indices = LTTBDownsampler().downsample(np.array(df["timestamp"]), np.array(df["value"]), n_out=3) was the fix

[jvdd]

Appreciate you contributing @jayceslesar 🚀

I haven't had the chance to fully explore the hyperplot project yet, but it's definitely on my radar.

I'll further comment on the UDF/pipe & polars integration suggestions in #30