fix(newos4): connect to Ollama on first use, honest write_file, list_files sees /sys

Three defects, each re-probed first and covered by a test that is red on HEAD.

- ollama-manager.js: nothing connects at boot; only the Ollama tab button called connect(),
  so a user's first message died with "Ollama not connected. Use ollamaconnect first." even
  when the endpoint was healthy. New _ensureConnected() runs from chat(), chatStream() and
  chatWithTools(): one attempt, shared by concurrent callers, a "Connecting…" hint, and only
  then the error — which now says the endpoint is unreachable instead of telling the user to
  run a command they already expected to work.
- module-aiassist.js write_file tool: fs.writeFile()'s false return was dropped and the tool
  answered "File written: …" anyway, sending the model on to a read that found nothing
  (MAINT-012). It now returns "Write FAILED: … — parent directory missing or target is a
  directory; nothing was written".
- module-aiassist.js list_files tool: the hand-rolled walk over root.content never resolved
  the virtual mounts, so /sys — a directory the UI and `ls` both show — answered
  "Directory not found" (MAINT-011). It now uses fs.readDir(), which resolves mounts and
  returns null for a genuinely missing path.

Tests: modules/test/test-ai-fs-tools.js (7 asserts) and
modules/test/test-ollama-autoconnect.js (11 asserts). RED on HEAD: ai tools 4 passed/3 FAILED
(/sys → "Directory not found", write_file → "File written: …"), ollama 2 passed/9 FAILED
("Use ollamaconnect first"). GREEN now 7/7 and 11/11; maintain --fix reports "node tests 9/9 pass".
Live on v4.6.105: connectedBefore=false, _ensureConnected()=true in 41 ms, 37 models.

1 files changed, 40 insertions, 6 deletions

diff --git a/modules/core/ollama-manager.js b/modules/core/ollama-manager.js
index 4f47db2..3d157ba 100644
--- a/modules/core/ollama-manager.js
+++ b/modules/core/ollama-manager.js
@@ -369,11 +369,36 @@ class OllamaManager {
         }
     }
 
+    // ─── Lazy connect ────────────────────────────────────────────────────────
+    /**
+     * Every request path calls this first. Nothing connects at boot, so the user's first message
+     * used to die with "Use ollamaconnect first" even when the endpoint was healthy
+     * (reported 2026-10-26/27). One attempt, then the caller prints the real reason.
+     * Concurrent callers share the same in-flight promise.
+     */
+    async _ensureConnected() {
+        if (this.connected) return true;
+        if (this._connecting) {
+            try { await this._connecting; } catch (e) { /* fall through to the state check */ }
+            return this.connected;
+        }
+        this.terminal.print('Connecting to Ollama…', 'info');
+        this._connecting = this.connect();
+        try { await this._connecting; } catch (e) {
+            this.terminal.log('WARN', 'Ollama', 'Auto-connect failed: ' + e.message);
+        }
+        this._connecting = null;
+        if (!this.connected) {
+            this.terminal.print('Auto-connect failed — check the endpoint in the Ollama tab (or run ollamaconnect).', 'warning');
+        }
+        return this.connected;
+    }
+
     // ─── Chat (non-streaming) ────────────────────────────────────────────────
 
     async chat(prompt, model = null) {
-        if (!this.connected) {
-            this.terminal.print('Ollama not connected. Use ollamaconnect first.', 'error');
+        if (!this.connected && !(await this._ensureConnected())) {
+            this.terminal.print('Ollama not connected: the endpoint is unreachable. Check the Ollama tab or run ollamaconnect.', 'error');
             return null;
         }
         const useModel = model || this.currentModel;
@@ -540,7 +565,10 @@ class OllamaManager {
     // ─── Chat Stream ─────────────────────────────────────────────────────────
 
     async chatStream(prompt, onToken, onComplete, model = null) {
-        if (!this.connected) { this.terminal.print('Ollama not connected.', 'error'); onComplete && onComplete(null); return; }
+        if (!this.connected && !(await this._ensureConnected())) {
+            this.terminal.print('Ollama not connected: the endpoint is unreachable.', 'error');
+            onComplete && onComplete(null); return;
+        }
         const useModel = model || this.currentModel;
         if (!useModel) { this.terminal.print('No model selected.', 'error'); onComplete && onComplete(null); return; }
 
@@ -711,8 +739,8 @@ class OllamaManager {
     }
 
     async chatWithTools(messages, tools, model = null) {
-        if (!this.connected) {
-            this.terminal.print('Ollama not connected. Use ollamaconnect first.', 'error');
+        if (!this.connected && !(await this._ensureConnected())) {
+            this.terminal.print('Ollama not connected: the endpoint is unreachable. Check the Ollama tab or run ollamaconnect.', 'error');
             return { message: null, tool_calls: null, error: 'Not connected' };
         }
         const useModel = model || this.currentModel;
@@ -893,7 +921,13 @@ class OllamaManager {
                 }
 
                 return {
-                    message: message ? { role: message.role || 'assistant', content: content } : null,
+                    message: message ? {
+                        role: message.role || 'assistant',
+                        content: content,
+                        // OpenAI-compatible providers require tool_calls INSIDE the assistant message;
+                        // a bare { role:'tool' } with no preceding tool_calls is rejected with HTTP 400.
+                        ...(Array.isArray(toolCalls) && toolCalls.length ? { tool_calls: toolCalls } : {}),
+                    } : null,
                     tool_calls: toolCalls ? toolCalls.map(tc => ({
                         id: tc.id,
                         type: tc.type || 'function',