fix: one tool-set recipe per session, so a tool cannot vanish between rounds

Two "unknown tool (not in this turn's tool set)" failures, one disease: the
turn's tool set was rebuilt from a different recipe depending on which entry
point happened to drive it.

A sub-agent got `ask_user_clarification`, `execute_subtask` and `activate_tools`
and nothing else — while `agents/common/tools.md` and every reporting agent's
prompt tell it to register its output with `update_scratchpad`. The child could
see the scratchpad injected into its context but had no way to write to it. It
now gets the scratchpad and todos tools, on the parent's `scratchpad_sid`: one
blackboard per session, as the surrounding code already declared.

`show_file_to_user` was injected per message by the WS handler, while
`resume_session` and `resolve_pending_call` rebuilt the list with `execute_task`
alone. So approving a card, or reconnecting mid-turn, continued the *same*
conversation with the tool silently gone. There is now a single recipe,
`ChatHub::session_interface_tools`, used by all three paths and fed by a builder
the shell installs once through `Skald::set_interface_tools_builder`: the core
keeps owning the tool, the shell keeps owning the policy of who gets it —
Telegram still does not, since it cannot act on OpenFile.

crates/skald-core/src/loop_adapters/catalog.rs

@@ -26,7 +26,9 @@ use crate::clarification::ClarificationManager;
 use crate::llm::LlmManager;
 use crate::llm::logging::RequestLogTarget;
 use crate::loop_adapters::activation::SkaldToolActivator;
-use crate::loop_adapters::builtins::{SkaldAskUserTool, SkaldHumanChannel};
+use crate::loop_adapters::builtins::{
+    SkaldAskUserTool, SkaldHumanChannel, UpdateScratchpadTool, WriteTodosTool,
+};
 use crate::loop_adapters::history::SqliteHistory;
 use crate::loop_adapters::prefix_cache::PrefixCache;
 use crate::loop_adapters::runtime::LoopConfig;
@@ -217,6 +219,18 @@ impl AgentCatalog for SkaldAgentCatalog {
             scope.session_id,
             Some(child_frame.get()),
         )))));
+        // The blackboard and the checklist. Both are *told* to sub-agents by the
+        // prompts (`agents/common/tools.md`, and every reporting agent ends with
+        // "register your report with `update_scratchpad`"), and the scratchpad is
+        // injected into a child's context — so leaving the writers out of the
+        // child's tool set made a documented instruction unexecutable: the model
+        // called them and got "unknown tool". `scratchpad_sid` is the parent's,
+        // deliberately (see the context above): one blackboard per session.
+        native.push(Arc::new(UpdateScratchpadTool::new(
+            self.pool.clone(),
+            scope.scratchpad_sid,
+        )));
+        native.push(Arc::new(WriteTodosTool));
 
         let toolset: Arc<dyn agent_loop::tool::ToolSet> = Arc::new(
             SkaldToolSet::new(