RESEARCH: Paper claims benchmarks that don't exist yet

[Manideep3969]

Issue

docs/notes/06-paper-outline.md Section 1 states specific numerical claims (1.3–2.8× depth reduction, 15–40% fidelity improvement, 3.5× throughput improvement) but no experimental validation has been conducted. These appear to be projected targets, not measured results.

Additionally:

• Section 8.3 describes a constraint satisfaction formulation for coherence-aware scheduling, but the implementation just sorts by priority (and has the ALAP bug — see BUG: ALAP schedule is identical to ASAP — not actually late-as-possible #32)
• Section 5.2 describes two-qubit gate fusion with single-qubit absorption, which is not implemented
• Section 4.2 includes crosstalk errors in the cost model, but the implementation has no crosstalk modeling

Fix

1. Remove specific numerical claims until experiments are completed, or label them as 'target' results
2. Align the paper description with the actual implementation (heuristic sort, not constraint satisfaction)
3. Remove the two-qubit gate fusion description or implement it
4. Remove crosstalk from the cost model formulation or add implementation