Keep the pixel forensics in the library, drop the ai-score tooling

`scripts/ai_score.py` and the dataset scanner that fed it are gone: the detector
they trained is not something this project runs, and the corpus lived outside the
repository anyway. Nothing else referenced them.

The scanner's pixel layer was worth keeping, so it moves into the package as
`pixel_evidence.py` -- six families of scale-robust statistics (block-DCT histograms
and Benford deviation, FFT band energies and CFA peaks, high-pass residual, error
level, gradient, colour) measured in a single shared decode. The arithmetic was
verified against the scanner over 60 corpus images, families and artifacts alike,
before the scanner was removed; that comparison is no longer possible, which is why
the tests now pin behavior instead: determinism, empty-not-wrong on images too small
for a family, and one failing family not taking the others with it.

It has no consumer. Nothing in the package reads it, and the module says so.

`artifacts=True` returns the spatial layer -- perceptual hash, 128px thumbnail,
coarse ELA/residual/phase maps. Those identify the source image rather than describe
it, so they are opt-in and separate: everything else is a scalar or a fixed-length
histogram nothing can be reconstructed from.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

docs/module-internals.md

@@ -425,6 +425,21 @@ Pixel forensics are deliberately absent: nothing in the provenance path reads th
 Verdict equivalence is checked over tracked fixtures and a separate local evaluation
 corpus.
 
+### Experimental pixel forensics
+
+[`pixel_evidence.py`](../src/remove_ai_watermarks/pixel_evidence.py) measures six
+families of scale-robust pixel statistics (block-DCT histograms and Benford
+deviation, FFT band energies and CFA peaks, high-pass residual, error level,
+gradient, colour) in a single decode, sharing the intermediate maps between them.
+
+It has no consumer. Nothing in the package reads it -- not the verdict, not removal,
+not the CLI -- and the shape is unstable until something does.
+
+`artifacts=True` additionally returns the spatial layer: a perceptual hash, a 128px
+JPEG thumbnail, and coarse ELA, residual and phase maps. Those identify the source
+image rather than describe it, which is why they are opt-in and a separate field: a
+caller storing them is handling image content, not statistics about it.
+
 The DWT-DCT detector and the visible-mark stage share a single decode of the
 source, held by
 a per-call `_SharedDecode`. It exposes two accessors because the two arms need