Fix review-cycle threshold wording inconsistencies in loop termination and Phase 4a entry condition

[williamthorsen]

Problem

Two minor text inconsistencies remain in packages/agents/content/skills/orchestrate/modules/review-cycle.md after the mode system and two-threshold model implementation (CODY-115).

Details

1. Loop-termination none wording (line ~187)

The convergence description lists none as a separate condition alongside "below both thresholds":

Repeat until convergence (aggregated criticality is none, or below both thresholds, ...)

This is inconsistent with the flow-control rules that absorbed none into the criticality < budget_threshold case. Since none is always below any threshold, listing it separately is redundant and could mislead a reader into thinking it's a distinct code path.

2. Phase 4a entry condition clause (line ~212)

The second clause in the Phase 4a trigger condition ("after fix cycles reduce criticality below the approval threshold") may be misleading. If budget remains and criticality is still >= budget_threshold, the engine continues opportunistic fixes — it doesn't exit to Phase 4a. The third clause already covers the correct terminal path (budget exhaustion). The second clause is either redundant or describes a scenario that can't actually trigger Phase 4a entry.

Acceptance criteria

• Loop-termination text removes redundant none mention or integrates it into the threshold-relative language
• Phase 4a entry condition accurately describes all paths that lead to Phase 4a, without implying paths that don't

Context

These were identified during the CODY-115 orchestrated review cycle (round 3 of 3, aspect-code-reviewer). Both are documentation polish with no behavioral impact — the flow-control rules themselves are correct.

Relates to #115.