fix: prevent agent from stopping mid-task — compression floor, budget overhaul, activity tracking

Three root causes of the 'agent stops mid-task' gateway bug:

1. Compression threshold floor (64K tokens minimum)
   - The 50% threshold on a 100K-context model fired at 50K tokens,
     causing premature compression that made models lose track of
     multi-step plans.  Now threshold_tokens = max(50% * context, 64K).
   - Models with <64K context are rejected at startup with a clear error.

2. Budget warning removal — grace call instead
   - Removed the 70%/90% iteration budget warnings entirely.  These
     injected '[BUDGET WARNING: Provide your final response NOW]' into
     tool results, causing models to abandon complex tasks prematurely.
   - Now: no warnings during normal execution.  When the budget is
     actually exhausted (90/90), inject a user message asking the model
     to summarise, allow one grace API call, and only then fall back
     to _handle_max_iterations.

3. Activity touches during long terminal execution
   - _wait_for_process polls every 0.2s but never reported activity.
     The gateway's inactivity timeout (default 1800s) would fire during
     long-running commands that appeared 'idle.'
   - Now: thread-local activity callback fires every 10s during the
     poll loop, keeping the gateway's activity tracker alive.
   - Agent wires _touch_activity into the callback before each tool call.

Also: docs update noting 64K minimum context requirement.

Closes #7915 (root cause was agent-loop termination, not Weixin delivery limits).

agent/model_metadata.py

@@ -85,6 +85,11 @@ CONTEXT_PROBE_TIERS = [
 # Default context length when no detection method succeeds.
 DEFAULT_FALLBACK_CONTEXT = CONTEXT_PROBE_TIERS[0]
 
+# Minimum context length required to run Hermes Agent.  Models with fewer
+# tokens cannot maintain enough working memory for tool-calling workflows.
+# Sessions, model switches, and cron jobs should reject models below this.
+MINIMUM_CONTEXT_LENGTH = 64_000
+
 # Thin fallback defaults — only broad model family patterns.
 # These fire only when provider is unknown AND models.dev/OpenRouter/Anthropic
 # all miss. Replaced the previous 80+ entry dict.