[None][test] CBTS coverage: log immediate-save failures instead of swallowing

The poller's post-context-switch save is the only persistence chance for a
short-lived worker that exits before the next periodic save, so a silent
`except: pass` turned lost subprocess coverage into an undebuggable false
negative. Print to stderr like the periodic/final save handlers do.

Signed-off-by: Ivy Zhang <25222398+crazydemo@users.noreply.github.com>

Author: crazydemo

jenkins/scripts/cbts/coverage_utils/sitecustomize.py

@@ -155,8 +155,11 @@ def _poll_marker():
                             try:
                                 # Save now so short-lived workers persist context before the periodic save.
                                 cov.save()
-                            except Exception:
-                                pass
+                            except Exception as e:
+                                print(
+                                    f"[cbts] immediate save failed in pid {os.getpid()}: {e!r}",
+                                    file=sys.stderr,
+                                )
                         last_seen = nodeid
                 except (FileNotFoundError, OSError):
                     pass