feat(model): Codex GPT slugs default back to 272K; explicit -900k picker variants opt into the verified large window

The Aug 16 change that auto-raised gpt-5.4/5.6 Codex OAuth context to the
live-verified 900K burned through subscription usage for users who never
asked for the larger window (bigger window = more input tokens per request).

- Base Codex slugs (gpt-5.6-sol/terra/luna, gpt-5.4) now resolve to the
  advertised 272K again — the cheaper limit is the default.
- The model picker synthesizes explicit <slug>-900k variants (e.g.
  gpt-5.6-sol-900k) for every live-verified slug; selecting one opts into
  the 900K window. Slugs that genuinely enforce 272K (gpt-5.5,
  gpt-5.4-mini) get no variant.
- The -900k suffix is Hermes-side only: stripped before the model id hits
  the wire (main transport + auxiliary Responses adapter), and pricing
  aliases the variants onto the base entries.
- Docs: new opt-in section in context-compression-and-caching.md.

agent/usage_pricing.py

@@ -994,11 +994,16 @@ _OFFICIAL_DOCS_PRICING: Dict[tuple[str, str], PricingEntry] = {
 
 # GPT-5.6 "-pro" high-effort variants bill at the same per-token rates as
 # their base tiers (more tokens per task, not a higher rate). Alias them
-# onto the base entries so the snapshot stays single-source.
+# onto the base entries so the snapshot stays single-source. The Hermes-side
+# "-900k" large-context Codex picker variants are the same underlying model
+# (the suffix is stripped on the wire), so they alias identically.
 for _base_56 in ("gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"):
     _OFFICIAL_DOCS_PRICING[("openai", f"{_base_56}-pro")] = _OFFICIAL_DOCS_PRICING[
         ("openai", _base_56)
     ]
+    _OFFICIAL_DOCS_PRICING[("openai", f"{_base_56}-900k")] = _OFFICIAL_DOCS_PRICING[
+        ("openai", _base_56)
+    ]
 del _base_56
 
 # The direct Gemini provider currently exposes preview IDs for these two