Data class constructor does not save the procnames value

Start in pyspi/data.py at Data.init and the procnames() property referenced by the issue. Run the provided reproduction to confirm the supplied names are ignored; done means Data.procnames and the Calculator table use the requested process names.

When calling the Data constructor Data.__init__, the constructor accepts procnames as a value, which it then validates:
https://github.com/DynamicsAndNeuralSystems/pyspi/blob/46ccfd522488ea518011c162f1e550ba4a1a013c/pyspi/data.py#L82-L83
but it never save the value, which it then attempts to read in the procnames() function:
https://github.com/DynamicsAndNeuralSystems/pyspi/blob/46ccfd522488ea518011c162f1e550ba4a1a013c/pyspi/data.py#L100-L106
Since self._procnames is never set procnames will always return the generated names.

The issue can be seen when running the following code(Generated by Claude Opus 5):

import inspect
import platform
import sys
from importlib.metadata import PackageNotFoundError, version

import numpy as np
from pyspi.calculator import Calculator
from pyspi.data import Data


def pkg_version(name: str) -> str:
    try:
        return version(name)
    except PackageNotFoundError:
        return "not installed"


print("--- environment")
print(f"python : {sys.version.split()[0]} ({platform.platform()})")
for pkg in ("pyspi", "numpy", "pandas"):
    print(f"{pkg:<7}: {pkg_version(pkg)}")
print(f"Data signature: {inspect.signature(Data)}")

# 3 processes x 100 observations, rows are processes (dim_order="ps")
X = np.random.default_rng(0).normal(size=(3, 100))
names = ["wrist", "thumb", "index"]

print("\n--- 1. Data object")
data = Data(data=X, dim_order="ps", procnames=names)
print(f"expected procnames: {names}")
print(f"actual procnames  : {list(data.procnames)}")
data_ok = list(data.procnames) == names

print("\n--- 2. Calculator table")
calc = Calculator(dataset=data, subset="fabfour")
calc.compute()
spis = list(dict.fromkeys(calc.table.columns.get_level_values(0)))
spi = "cov_EmpiricalCovariance" if "cov_EmpiricalCovariance" in spis else spis[0]
print(f"calc.table[{spi!r}]:")
print(calc.table[spi])
table_ok = list(calc.table.index) == names

print("\n--- result")
print(f"Data.procnames uses given names : {data_ok}")
print(f"Calculator.table uses given names: {table_ok}")
if data_ok and table_ok:
    print("Not reproduced: procnames were applied.")
    sys.exit(0)
print("Reproduced: procnames passed to Data are not applied.")
sys.exit(1)

It produces the following output:

--- environment
python : 3.12.10 (Windows-11-10.0.26100-SP0)
pyspi  : 2.0.1
numpy  : 1.26.4
pandas : 2.3.3
Data signature: (data=None, dim_order='ps', detrend=False, normalise=True, name=None, procnames=None, n_processes=None, n_observations=None)

--- 1. Data object
[1/2] Skipping detrending of time series in the dataset...
[2/2] Normalising (z-scoring) each time series in the dataset...

expected procnames: ['wrist', 'thumb', 'index']
actual procnames  : ['proc-0', 'proc-1', 'proc-2']

--- 2. Calculator table
calc.table['cov_EmpiricalCovariance']:
process    proc-0    proc-1    proc-2
proc-0        NaN  0.054899 -0.013441
proc-1   0.054899       NaN  0.030280
proc-2  -0.013441  0.030280       NaN

--- result
Data.procnames uses given names : False
Calculator.table uses given names: False
Reproduced: procnames passed to Data are not applied.