Bash – How to repeat a command with an added parameter

There is no issue with the \n. This is yet again the old escape sequence length problem: \e[0m and similar do not contribute to the actual length of the prompt, so you have to enclose them in \[..\] to indicate this to the interpreter:

PS1="\[\e[0;36m\]\h\[\e[m\] \[\e[0;33m\]\w/\[\e[m\]\n \[\e[0;31m\]\$ →\[\e[m\]  "

You can use HISTIGNORE environment variable.

A colon-separated list of patterns used to decide which command lines should be saved on the history list. For example:

export HISTIGNORE="cd*:ls*"

this will ignore all cd and ls commands from history.

EDIT:

sorry, In my first reading I didn't take sufficent care to your question. Btw, you are right: HISTIGNOREis a very hard setting, and bash_history will simply ignore each command that is listed in this variable (it will ignore it at the moment you execute it, if there are already entries with these command into your HISTFILE they will be found by history reverse search.

To first answer to your last question, live and written history may vary a lot if you use different terminals simultaneously. You could try a solution as described here:

Preserve bash history in multiple terminal windows

What you could imagine is a solution in which you use a function in your ~/.bash_logout file, which will erase the history entries that you do not want.

something like this should do what you want:

 function clear_history(){
    # hist_ignore regex, this may be done in a more fancy way, 
    # such as using an env variable, but anyway
    MYHISTIGNORE='^cd:^view:^cat'

    # write down history to file
    history -w

    # create a new tmp file for history
    tmpFile=$(mktemp)
    cp ${HISTFILE}  ${tmpFile}

    # use it to filter history
    for myRegex in $(echo $MYHISTIGNORE | awk 'BEGIN{FS=":"}{for (i=1;i<=NF;i++){print $i}}'); do 
       sed -i "/${myRegex}/d" ${tmpFile}
    done

    # and copy it back to history file
    mv ${tmpFile} ${HISTFILE}
}