refactor(agents): move mac-only modules out of the cross-agent shared kernel

app/agents/shared/ is a sibling of anonymous_chat/podcaster/multi_agent_chat/
video_presentation, so it should only hold code shared across 2+ of those
agents. In practice podcaster and video_presentation import nothing from it,
and anonymous_chat needs only context + compaction + retry_after + web_search.
Everything else was multi_agent_chat-only (the boundary just passes through).

Move the multi_agent_chat-only cluster into multi_agent_chat/shared/ (files
moved verbatim via git rename; ~116 import sites rewritten):

  errors, feature_flags, filesystem_selection, path_resolver, prompt_caching,
  sandbox, llm_config, mention_resolver
  middleware/busy_mutex, middleware/kb_persistence

busy_mutex/llm_config/mention_resolver are boundary-only but import the moved
modules, so they were folded in to avoid a backwards shared -> multi_agent_chat
dependency. main_agent builders now import the impls directly; the shared
middleware barrel keeps only the genuinely-shared compaction + retry_after.

Also delete the dead leftover shared/plugins and shared/skills dirs (live
copies already live under main_agent/).

Remaining in app/agents/shared/: context, system_prompt(+prompts), checkpointer,
middleware/{compaction,retry_after,dedup_tool_calls}, tools/. checkpointer and
system_prompt are boundary-only infra pending a dedicated home decision.

surfsense_backend/tests/unit/services/test_supports_image_input.py

@@ -227,7 +227,7 @@ global_llm_configs:
 
 
 def test_agent_config_from_yaml_explicit_overrides_resolver():
-    from app.agents.shared.llm_config import AgentConfig
+    from app.agents.multi_agent_chat.shared.llm_config import AgentConfig
 
     cfg_text_only = AgentConfig.from_yaml_config(
         {
@@ -256,7 +256,7 @@ def test_agent_config_from_yaml_explicit_overrides_resolver():
 def test_agent_config_from_yaml_unannotated_uses_resolver():
     """Without an explicit YAML key, AgentConfig defers to the catalog
     resolver — for ``gpt-4o`` LiteLLM's map says supports_vision=True."""
-    from app.agents.shared.llm_config import AgentConfig
+    from app.agents.multi_agent_chat.shared.llm_config import AgentConfig
 
     cfg = AgentConfig.from_yaml_config(
         {
@@ -275,7 +275,7 @@ def test_agent_config_auto_mode_supports_image_input():
     so users can keep their selection on Auto with a vision-capable
     deployment somewhere in the pool. The router's own `allowed_fails`
     handles non-vision deployments via fallback."""
-    from app.agents.shared.llm_config import AgentConfig
+    from app.agents.multi_agent_chat.shared.llm_config import AgentConfig
 
     auto = AgentConfig.from_auto_mode()
     assert auto.supports_image_input is True

surfsense_backend/tests/unit/services/test_vision_llm_api_base_defense.py

@@ -61,7 +61,7 @@ async def test_get_vision_llm_global_openrouter_sets_api_base():
             return_value=cfg,
         ),
         patch(
-            "app.agents.shared.llm_config.SanitizedChatLiteLLM",
+            "app.agents.multi_agent_chat.shared.llm_config.SanitizedChatLiteLLM",
             new=FakeSanitized,
         ),
     ):