[FIX] tests: stop test runner from opening error dialogs when not ready

Previously, the test runner would evaluate an expression to check if the
browser test that is about to be run is ready, but before the test is
ready, this expression may be invalid as the variables used in the
expression may not be defined yet, causing a ReferenceError to be thrown
by Chrome.

In Chrome >=102, errors that are thrown when writing code in the console
or by using Runtime.evaluate over CDP are thrown in the context of the
current tab, which means that they trip registered error handlers in
that tab. In Odoo, this means that we show error dialogs with the
traceback.

In the tour manager, when we are looking for an element  to trigger, we
only look for that element inside dialogs if there are any dialogs open
(unless the in_dialog option is false on that specific step). This means
that if an error dialog is open, most tours will fail (which is actually
what we want).

In order to avoid opening a bunch of error dialogs while waiting for the
tour to be ready, we simply wrap the ready expression in a try catch so
that it doesn't throw an error, and simply returns a undefined until the
tour is ready instead of throwing a ReferenceError.

closes odoo/odoo#95635

X-original-commit: 3872dbd63233d4c6d961a3cd896b84e4301a1f84
Signed-off-by: Xavier Morel (xmo) <xmo@odoo.com>
Signed-off-by: Samuel Degueldre <sad@odoo.com>

odoo/tests/common.py

@@ -1342,7 +1342,7 @@ class ChromeBrowser:
                 break
 
             result = self._websocket_request('Runtime.evaluate', params={
-                'expression': ready_code,
+                'expression': "try { %s } catch {}" % ready_code,
                 'awaitPromise': True,
             }, timeout=timeout-taken)['result']