This update for crmsh fixes the following issues:

- Update to version 5.0.0+20260403.7274d51c:
    * Fix: bootstrap: Ensure robust node identification when removing from cluster (bsc#1259683)
    * Fix: utils: Raise UnreachableNodeError for those ssh unreachable nodes (bsc#1250645)
    * Fix: sbd: Update the formular to calculate the expected fencing-watchdog-timeout
    * Fix: log: Disable color when not on a TTY (bsc#1259178)
    * Fix: cibverify: If no errors found, treat crm_verify result as success (bsc#1250349)
    * Fix: log: Add milliseconds time format to crmsh.log (bsc#1255021)
    * Fix: ui_cluster: Stop dlm in maintenance mode correctly (bsc#1253733)
    * Dev: utils: Rename utils.ssh_reachable_check to utils.ssh_port_reachable_check
    * Dev: utils: Check if the peer node needs password to access
    * Dev: utils: Rename utils.node_reachable_check to utils.ssh_reachable_check
    * Dev: utils: Adjust node_reachable_check function and the way it is used
    * Dev: ui_utils: Move ui_node.parse_option_for_nodes to ui_utils.parse_and_validate_node_args
    * Dev: doc: Update the formular to calculate the expected fencing-watchdog-timeout
    * Dev: bootstrap: Sync directory sbd.SBDManager.SBD_SYSTEMD_DELAY_START_DIR
    * Dev: bootstrap: Refactor retrieve_all_config_files function
    * Dev: behave_agent: Prevent client hang behind firewall during long cluster joins
    * Dev: sh: Use sh_helper.py for su commands (bsc#1254757)
    * Dev: doc: Mention about watchdog-device option also acceptes driver name
    * Dev: watchdog: Improve the fatal error logging message
    * Dev: ui_cluster: Hint the watchdog option should be used with sbd option
    * Dev: bootstrap: Skip inactive cluster node when calling restart_cluster function
    * Dev: corosync: Add milliseconds to log time format
    * Dev: utils: Reuse methods in xmlutil.CrmMonXmlParser