feat: wire context engine plugin slot into agent and plugin system

- PluginContext.register_context_engine() lets plugins replace the
  built-in ContextCompressor with a custom ContextEngine implementation
- PluginManager stores the registered engine; only one allowed
- run_agent.py checks for a plugin engine at init before falling back
  to the default ContextCompressor
- reset_session_state() now calls engine.on_session_reset() instead of
  poking internal attributes directly
- ContextCompressor.on_session_reset() handles its own internals
  (_context_probed, _previous_summary, etc.)
- 19 new tests covering ABC contract, defaults, plugin slot registration,
  rejection of duplicates/non-engines, and compressor reset behavior
- All 34 existing compressor tests pass unchanged

run_agent.py

@@ -1268,19 +1268,32 @@ class AIAgent:
                                         pass
                         break
         
-        self.context_compressor = ContextCompressor(
-            model=self.model,
-            threshold_percent=compression_threshold,
-            protect_first_n=3,
-            protect_last_n=compression_protect_last,
-            summary_target_ratio=compression_target_ratio,
-            summary_model_override=compression_summary_model,
-            quiet_mode=self.quiet_mode,
-            base_url=self.base_url,
-            api_key=getattr(self, "api_key", ""),
-            config_context_length=_config_context_length,
-            provider=self.provider,
-        )
+        # Check if a plugin registered a custom context engine (e.g. LCM)
+        _plugin_engine = None
+        try:
+            from hermes_cli.plugins import get_plugin_context_engine
+            _plugin_engine = get_plugin_context_engine()
+        except Exception:
+            pass
+
+        if _plugin_engine is not None:
+            self.context_compressor = _plugin_engine
+            if not self.quiet_mode:
+                logger.info("Using plugin context engine: %s", _plugin_engine.name)
+        else:
+            self.context_compressor = ContextCompressor(
+                model=self.model,
+                threshold_percent=compression_threshold,
+                protect_first_n=3,
+                protect_last_n=compression_protect_last,
+                summary_target_ratio=compression_target_ratio,
+                summary_model_override=compression_summary_model,
+                quiet_mode=self.quiet_mode,
+                base_url=self.base_url,
+                api_key=getattr(self, "api_key", ""),
+                config_context_length=_config_context_length,
+                provider=self.provider,
+            )
         self.compression_enabled = compression_enabled
         self._subdirectory_hints = SubdirectoryHintTracker(
             working_dir=os.getenv("TERMINAL_CWD") or None,
@@ -1397,15 +1410,9 @@ class AIAgent:
         # Turn counter (added after reset_session_state was first written — #2635)
         self._user_turn_count = 0
 
-        # Context compressor internal counters (if present)
+        # Context engine reset (works for both built-in compressor and plugins)
         if hasattr(self, "context_compressor") and self.context_compressor:
-            self.context_compressor.last_prompt_tokens = 0
-            self.context_compressor.last_completion_tokens = 0
-            self.context_compressor.compression_count = 0
-            self.context_compressor._context_probed = False
-            self.context_compressor._context_probe_persistable = False
-            # Iterative summary from previous session must not bleed into new one (#2635)
-            self.context_compressor._previous_summary = None
+            self.context_compressor.on_session_reset()
     
     def switch_model(self, new_model, new_provider, api_key='', base_url='', api_mode=''):
         """Switch the model/provider in-place for a live agent.