v10.22.0 — sharper N=6 diagnosis (P1 follow-up)

The N=6 case (a/N=1/6,5/6) remains OPEN, but the reason is now precise: the q=i reflected-contour argument that closes N=4 despite tie height T<0 relies on a ray lemma needing sin(theta)!=0 at the tie-ray angle. N=6's certified tie-ray angle is exactly theta*=0, degenerating that lemma's proof exactly where N=6 needs it. Two alternative hypotheses (a 3-way tie, a composite-divisor obstruction) were checked and ruled out. This is a diagnosis, not an impossibility proof: a modulus-only rescue at theta=0 was not checked and remains conceivable. Reviewed by Reviewer AK (CORRECT, one wording fix).