[BUG] Ensure gamma/poission always get positive y in doctest

[Omswastik-11]

Fixes #659

Issue

---

The issue was in doc_test itwas passing non-positive values to distributions like poisson and binomial

Exception raised:
Traceback (most recent call last):
File "C:\Users\omswa\AppData\Local\Programs\Python\Python312\Lib\doctest.py", line 1368, in __run
exec(compile(example.source, filename, "single",
File "<doctest GAMRegressor[11]>", line 1, in
gam_gamma.fit(X, y)
File "C:\Users\omswa\Desktop\skpro\skpro\regression\base_base.py", line 116, in fit
return self._fit(X_inner, y_inner)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\omswa\Desktop\skpro\skpro\regression\gam_gam.py", line 189, in fit
self.estimator.fit(X_np, y_np)
File "C:\Users\omswa\Desktop\skpro.venv\Lib\site-packages\pygam\pygam.py", line 886, in fit
y = check_y(y, self.link, self.distribution, verbose=self.verbose)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\omswa\Desktop\skpro.venv\Lib\site-packages\pygam\utils.py", line 238, in check_y
raise ValueError(
ValueError: y data is not in domain of log link function. Expected domain: [np.float64(0.0), np.float64(inf)], but found [-169.89, 312.82]

Approach

---

Ensured always positive values being passed to poission or gamma .

[fkiraly]

Ah, good catch