Describe the bug

When HidalgoSegmenter is configured with a high burn_in value (e.g., 0.9) and low sampling_rate (default 10), the filtering condition it % sampling_rate == 0 and it >= n_iter * burn_in may result in an empty idx array. This causes bestsampling to remain uninitialized, leading to an IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed when attempting to slice it as bestsampling[:, :K].

Root Cause:
The code doesn't validate that at least one sample passes the filtering criteria before attempting to access the sampled data.

Steps to Reproduce:
import numpy as np
from aeon.segmentation import HidalgoSegmenter

X = np.random.rand(50, 3)
model = HidalgoSegmenter(K=2, n_iter=10, sampling_rate=10, burn_in=0.9, seed=42)
model.fit(X) # IndexError raised

Expected Behavior: Should either produce valid samples or raise a clear error message indicating invalid parameter configuration.

Steps/Code to reproduce the bug

import numpy as np
from aeon.segmentation import HidalgoSegmenter

np.random.seed(42)
X = np.random.rand(50, 3)
model = HidalgoSegmenter(K=2, n_iter=10, sampling_rate=10, burn_in=0.9, seed=42)
model.fit(X)

Expected results

No error is thrown, or a clear ValueError message indicating invalid parameter configuration.

Actual results

Traceback (most recent call last):
  File "example_code.py", line 5, in <module>
    model.fit(X)
  File "aeon/segmentation/_hidalgo.py", line 315, in _fit
    self._d = np.mean(bestsampling[:, :K], axis=0)
IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed

Versions

aeon: 1.3.0+
numpy: 1.24+