# Bosonic Continuous Interpolation Experiment

Testing a unified Monte Carlo estimator without regime switching.

## Goal
Replace hard p≥5 cutoff with continuous λ(p) interpolation:
```
λ(p) = exp(-p²/N)
Q(p) = λ(p) Q_bos + (1-λ(p)) Q_fact
```

## Structure
- `kernel/ryser_continuous.py` — Unified permanent estimator
- `kernel/mode_sampler.py` — Mode sampling kernel
- `tests/test_lambda_interpolation.py` — Verify continuity

## Key Change
Remove:
```python
if p >= 5: distinguishable_k()
```

Replace with single estimator using Ryser permanent for all p ≤ 6.