warn when n_factors exceeds number of features

[Qayad-Ali]

Summary

FactorAnalyzer.fit caps the number of factors when n_factors is greater than the number of features — it returns a truncated loadings matrix with no indication to the user:

import numpy as np
from factor_analyzer import FactorAnalyzer

X = np.random.default_rng(0).normal(size=(200, 6))
fa = FactorAnalyzer(n_factors=10, rotation=None).fit(X)
print(fa.loadings_.shape)   # (6, 6) — 10 requested, silently reduced to 6, no warning

In factor analysis the number of factors can't exceed the number of observed variables, so requesting more is mostly always a mistake. This gives a UserWarning in that case — consistent with the existing warning when the number of factors is constrained by the sample size.

Changes

• FactorAnalyzer.fit: warn when n_factors > n_features.
• Added a test in tests/test_factor_analyzer.py.

No change to results for valid inputs.