feat(agent): make the assistant agentic - visible tools, LLM voice, full toolset

The agent felt like an artificial idiot because the LLM almost never spoke
for itself: 14+ Go paths injected fmt.Sprintf canned replies, the frontend
filtered out tool-progress events so users saw three dots for 10-20s, the
main prompt told the LLM "be a trading partner" AND "answer only what's
asked", and the planner sliced the toolset by inferred domain so a "BTC
dropped, how much am I losing?" question couldn't see positions and market
at the same time.

- agent/central_brain.go: shouldTrustDeterministicSkillReply now always
  returns false. Successful mutations (trader/strategy/model/exchange
  create/update/start/stop/delete) flow through reviewTaskCompletion so the
  LLM sees the real outcome JSON and writes the user-facing prose. The
  trade-confirmation regex path (handleTradeConfirmation) was already
  outside this code path and is unaffected.

- agent/agent.go: rewrite the Behavior section of the main system prompt.
  Replace the contradictory "answer only what's asked / don't upsell" with
  "lead with the direct answer, then optionally one relevant follow-up
  only when (a) open risk, (b) missing config, or (c) the next step is
  obvious — e.g. created, want me to start it?". Explicitly authorize
  chaining ("if the user says create and start, do both this turn") and
  ban "please wait / I'll get back to you" language because there is no
  background job to come back from.

- agent/tools.go: plannerToolsForText always returns the full 22-tool set
  (new __all__ domain). The old per-domain trimming hid manage_trader from
  market questions and execute_trade from anything that didn't look like
  an explicit trade — cross-domain reasoning was structurally blocked. The
  compact-vs-full strategy schema switch is preserved so mutation intents
  still see the full config schema.

- web/src/components/agent/{AgentStepPanel,ChatMessages}.tsx: stop
  filtering tool: steps. Map raw tool names to friendly labels with emoji
  ("get_positions" → "📊 检查持仓") in zh/en/id. Users now see what the
  agent is doing in real time instead of silence. central_brain routing
  chatter still gets dropped.

- agent/planner_tools_test.go: tests updated to assert the new
  full-toolset behavior and the compact-vs-full strategy schema switch.

agent/central_brain.go

@@ -937,17 +937,22 @@ func (a *Agent) executeActiveSkillSession(storeUserID string, userID int64, lang
 	return outcome, ActiveSkillSession{}, false, true
 }
 
-func shouldTrustDeterministicSkillReply(outcome skillOutcome) bool {
-	if outcome.Status != skillOutcomeSuccess || !outcome.GoalAchieved {
-		return false
-	}
-	switch outcome.Skill {
-	case "strategy_management", "trader_management", "model_management", "exchange_management":
-		switch outcome.Action {
-		case "create", "update", "update_name", "update_bindings", "configure_strategy", "configure_exchange", "configure_model", "update_status", "update_endpoint", "update_config", "update_prompt", "delete", "start", "stop", "activate", "duplicate":
-			return true
-		}
-	}
+// shouldTrustDeterministicSkillReply controls whether a Go-generated UserMessage
+// is shown verbatim to the user (true) or whether the LLM gets to review the
+// tool outcome and write a natural reply (false).
+//
+// Historically this returned true for every successful mutation on trader /
+// strategy / model / exchange — which meant the user always saw the same
+// canned `fmt.Sprintf` lines (e.g. "已创建 Trader: X. 我没有自动启动..."), and
+// the agent felt mechanical / "non-agentic". It now always returns false so the
+// LLM owns the voice. The cost is one extra LLM call per mutation; the upside
+// is that the agent can chain ("trader created — want me to start it now?"),
+// apologize on errors in plain language, respect the user's language and
+// tone, and behave like an actual agent instead of a settings panel.
+//
+// The trade-confirmation flow (execute_trade -> "确认 trade_xxx") is unaffected:
+// it runs through handleTradeConfirmation in trade.go before this code path.
+func shouldTrustDeterministicSkillReply(_ skillOutcome) bool {
 	return false
 }