fix(newos4): name the real Ollama failure, bound auto-connect, fix the review findings

Second pass over v4.6.105, driven by an independent review that reproduced each item.

- ollama-manager.js: chat() now reports the captured cause ("Ollama error: Failed to fetch",
  "…: HTTP 500") instead of the generic "No provider responded"; _lastChatError is reset per
  call so an old reason is never replayed. chatWithTools carries the reason in its error string.
- ollama-manager.js: _ensureConnected() bounds connect() with a deadline (connectTimeoutMs,
  default 10 s). connect() uses a bare fetch(), so a blackholed endpoint previously left the
  shared in-flight promise pending forever and hung every later message — the old code failed
  fast. The attempt is now try/finally, so a throwing terminal.log() can no longer poison
  _connecting and suppress the retry on the next message.
- ollama-manager.js: an automatic connect no longer silently rewrites the user's saved model;
  the previous choice is restored in memory and in localStorage, and the user is told.
- ollama-manager.js: the DeepSeek empty-content branch returned top-level tool_calls with
  content '', so the caller pushed an assistant message with no tool_calls and then a
  role:'tool' reply — an orphan tool message, which is the HTTP 400 that branch exists to
  prevent. tool_calls now live inside the message as well.
- module-aiassist.js: a no-argument list_files returned the ROOT instead of the cwd (readDir('/')
  instead of ls()); it now means the current directory again. Error text keeps the path as the
  caller wrote it, and write_file distinguishes a read-only virtual path (/sys/…) from a missing
  parent instead of blaming the parent.

Tests: test-ollama-error-text.js (new, 14 asserts, RED 9/5 before the edit), test-ai-fs-tools.js
7 → 16 asserts (write onto a directory, /sys write, list a file, '/', no-arg cwd, relative path),
test-ollama-autoconnect.js 11 → 19 (blackholed endpoint settles + retries, saved model survives;
the old "connecting hint" assertion is now falsifiable — the reviewer showed it also matched
connect()'s pre-existing line). Suite: 10/10 files green. Live on v4.6.106: healthy auto-connect
42 ms with the hint; blackholed endpoint settles in 1242 ms with connectTimeoutMs=1200, prints
"timed out" and "unreachable", _connecting cleared.

1 files changed, 33 insertions, 1 deletions

diff --git a/modules/test/test-ollama-autoconnect.js b/modules/test/test-ollama-autoconnect.js
index b9db4ff..12b0786 100644
--- a/modules/test/test-ollama-autoconnect.js
+++ b/modules/test/test-ollama-autoconnect.js
@@ -59,7 +59,7 @@ const json = (body, okFlag = true) => Promise.resolve({ ok: okFlag, status: okFl
         const res = await om.chat('ping');
         const text = out.map(o => o.text).join(' | ');
         ok('A1 connect is attempted lazily', seen.some(u => u.includes('/api/tags')), seen.join(', '));
-        ok('A2 the user sees a connecting hint', /Connecting to Ollama/.test(text), text.slice(0, 160));
+        ok('A2 the user sees a first-use hint', /connecting on first use/.test(text), text.slice(0, 160));
         ok('A3 manager is connected after the attempt', om.connected === true);
         ok('A4 the message is answered', String(res).includes('pong'), String(res).slice(0, 120));
         ok('A5 no "ollamaconnect first" error', !/ollamaconnect first/.test(text), text.slice(0, 200));
@@ -89,6 +89,38 @@ const json = (body, okFlag = true) => Promise.resolve({ ok: okFlag, status: okFl
         ok('C1 three concurrent messages probe /api/tags once', tagCalls === 1, 'tagCalls=' + tagCalls);
     }
 
+    // D. a blackholed endpoint must not hang every later message
+    {
+        const { om, out } = mk(() => new Promise(() => {}));   // never settles
+        om.connectTimeoutMs = 100;
+        const t0 = Date.now();
+        const res = await om.chat('ping');
+        const ms = Date.now() - t0;
+        const text = out.map(o => o.text).join(' | ');
+        ok('D1 chat settles on a silent endpoint', ms < 3000, ms + 'ms');
+        ok('D2 returns null', res === null, String(res));
+        ok('D3 the timeout is reported', /timed out/.test(text), text.slice(0, 220));
+        ok('D4 _connecting is cleared', om._connecting === null, String(om._connecting));
+        const before = out.length;
+        await om.chat('again');
+        ok('D5 the next message retries the connect', out.length > before, 'lines ' + before + ' -> ' + out.length);
+    }
+
+    // E. an automatic connect must not rewrite the model the user saved
+    {
+        store['webx11_ollama_model'] = 'my-saved-model';
+        const { om, out } = mk((url) => url.includes('/api/tags')
+            ? json({ models: [{ name: 'other-model' }] })
+            : json({ response: 'x', eval_count: 1, prompt_eval_count: 1 }));
+        om.currentModel = 'my-saved-model';
+        await om._ensureConnected();
+        const text = out.map(o => o.text).join(' | ');
+        ok('E1 the saved model survives an auto-connect', om.currentModel === 'my-saved-model', om.currentModel);
+        ok('E2 the stored preference is restored', store['webx11_ollama_model'] === 'my-saved-model', String(store['webx11_ollama_model']));
+        ok('E3 the user is told why', /Keeping your saved model/.test(text), text.slice(0, 200));
+        delete store['webx11_ollama_model'];
+    }
+
     if (fails === 0) {
         console.log('test-ollama-autoconnect: ' + passes + ' passed, 0 failed');
     } else {