SpamAssassin - do I need to unquote Bayes etc?

[flameproof]

SYSTEM INFORMATION
OS type and version	Debian 12
Virtualmin version	latest

Silly question: in Virtualmin the Bayesian learning classifier is set to default (means yes) - do I need to unquote it also in the local.cf?

#   Use Bayesian classifier (default: 1)
#  use_bayes 1

#   Bayesian classifier auto-learning (default: 1)
#  bayes_auto_learn 1

#   Set headers which may provide inappropriate cues to the Bayesian classifier
#
# bayes_ignore_header X-Bogosity
# bayes_ignore_header X-Spam-Flag
# bayes_ignore_header X-Spam-Status

Here an up to date SPAM header:

X-Spam-Flag: YES
X-Spam-Level: ********
X-Spam-Status: Yes, score=8.8 required=5.0 tests=DATE_IN_PAST_03_06,DMARC_NONE,
    HEADER_FROM_DIFFERENT_DOMAINS,HTML_MESSAGE,HTML_MIME_NO_HTML_TAG,
    HTML_OBFUSCATE_05_10,MIME_HTML_ONLY,RCVD_IN_SBL_CSS,
    RCVD_IN_VALIDITY_CERTIFIED_BLOCKED,RCVD_IN_VALIDITY_RPBL_BLOCKED,
    RCVD_IN_VALIDITY_SAFE_BLOCKED,SPF_HELO_NONE,SPF_NONE,URIBL_DBL_PHISH
    autolearn=no autolearn_force=no version=4.0.0

I am just wondering if learning is switched on or not. Seems I keep getting the same SPAM again and again. Additional, I have RoundCube and flag SPAM.

[flameproof]

Thank you. How can I check if RoundCube is properly flagging and reporting spam?

(Seems I had no Junk folder, I just created that in case that matters)