feat: wire CAI_SUPPORT_MODEL + CAI_SUPPORT_INTERVAL to auto-compact the main agent context

Problem
-------
CAI_SUPPORT_MODEL and CAI_SUPPORT_INTERVAL were documented environment
variables that had no runtime implementation.  The support/reasoner agent
was constructed using CAI_SUPPORT_MODEL but was never invoked automatically.
CAI_SUPPORT_INTERVAL existed only in docs/config tables with no scheduler
reading it.  As a result, users with a limited context window (e.g. 32k on a
local llama.cpp setup) had no way to automatically keep the main model's
message_history from overflowing during a long pentest.

Solution
--------
Added an auto-compact scheduler block immediately after turn_count += 1 in
the main single-agent run loop (run_cai_cli).

When both CAI_SUPPORT_MODEL and CAI_SUPPORT_INTERVAL are set the scheduler:

1. Fires every CAI_SUPPORT_INTERVAL turns (modulo check).
2. Calls COMPACT_COMMAND_INSTANCE._perform_compaction(model_override=
   CAI_SUPPORT_MODEL) which:
   - Sends the full message_history to the support model for summarisation.
   - Clears message_history entirely (hard context reset).
   - Saves the summary to /memory as a .md file.
   - Stores the summary in COMPACTED_SUMMARIES under the agent name.
3. Re-syncs the local agent variable from AGENT_MANAGER.get_active_agent()
   so the run loop continues with the freshly reloaded agent instance whose
   system prompt already contains the injected summary (the system prompt
   template calls get_compacted_summary() dynamically on every turn so no
   extra wiring was needed).
4. Prints a visible yellow/green indicator so users can see when compaction
   fires and confirm the context window has been reset.
5. Silently swallows errors (only logs when CAI_DEBUG=2) so a failing support
   model never crashes the main session.

Usage
-----
  CAI_SUPPORT_MODEL="openai/support"  # lighter model on litellm proxy
  CAI_SUPPORT_INTERVAL=4              # compact every 4 turns

src/cai/cli.py

@@ -1711,6 +1711,41 @@ def run_cai_cli(
                 agent.model.message_history[:] = fix_message_list(agent.model.message_history)
             turn_count += 1
 
+            # Auto-compact: when CAI_SUPPORT_MODEL + CAI_SUPPORT_INTERVAL are both set,
+            # compact the conversation every N turns using the support model so the
+            # main model's context window is kept small.  After compaction the
+            # summary is injected into the agent's system prompt and the local
+            # agent reference is refreshed so the loop uses the reloaded instance.
+            _support_model = os.getenv("CAI_SUPPORT_MODEL")
+            _support_interval_raw = os.getenv("CAI_SUPPORT_INTERVAL")
+            if _support_model and _support_interval_raw:
+                try:
+                    _support_interval = int(_support_interval_raw)
+                    if _support_interval > 0 and turn_count % _support_interval == 0:
+                        from cai.repl.commands.compact import COMPACT_COMMAND_INSTANCE
+                        console.print(
+                            f"\n[bold yellow]⟳ Auto-compact: turn {turn_count} "
+                            f"(every {_support_interval} turns) — "
+                            f"summarising with {_support_model}[/bold yellow]"
+                        )
+                        COMPACT_COMMAND_INSTANCE._perform_compaction(
+                            model_override=_support_model
+                        )
+                        # Re-sync the local agent reference so the loop continues
+                        # with the freshly reloaded agent (history cleared, memory
+                        # summary already injected into its system prompt).
+                        from cai.sdk.agents.simple_agent_manager import AGENT_MANAGER as _AM
+                        _reloaded = _AM.get_active_agent()
+                        if _reloaded is not None:
+                            agent = _reloaded
+                        console.print(
+                            "[bold green]✓ Memory summary applied to agent system prompt — "
+                            "context window reset[/bold green]\n"
+                        )
+                except (ValueError, Exception) as _e:
+                    if os.getenv("CAI_DEBUG", "1") == "2":
+                        console.print(f"[red]Auto-compact error: {_e}[/red]")
+
             # Stop measuring active time and start measuring idle time again
             stop_active_timer()
             start_idle_timer()