Add shellcheck CI check + address its concerns in few bash scripts found

[yarikoptic]

I (newbie to slackdump) was looking for incremental backup setup. Ran into bash script -- decided to contribute but also to robustify before I "mass try it" since shellcheck is right in warning (generally ;-) ) about such constructs as ( $(im_channels) ) since they might not handle correctly cases where names have spaces. I do not have proof yet though here that it would apply. Neither I have yet tested in this script that the helper append_to_array works.

but here is a script which demonstrates problem and checks different solutions for it

#!/bin/bash
# Disabling shellcheck check here since adding escaped symbols for demo
# shellcheck disable=SC2028

function func {
    echo word
    echo "phrase with space"
    echo 'wordwith\n'
}

function func2 {
    echo 'last-word\r'
}

function append_to_array {
    local array_name="$1"
    shift
    #local cmd="$@"

    # Read output of the command into a temporary array, splitting on newline
    # shellcheck disable=SC2034
    if [ -n "$*" ]; then
        IFS=$'\n' read -r -d '' -a temp_array < <("$@")
    else
        # no command provided -- read from stdin
        IFS=$'\n' read -r -d '' -a temp_array
    fi

    # Use eval to safely append temp_array to the named array
    eval "$array_name+=(\"\${temp_array[@]}\")"
}

function show {
    for i in "${LIST[@]}"; do
        echo "<$i>"
    done
}

echo "Can't use mapfile since not portable -- not available on OSX"

echo "Simple array: not good -- would not be good for spaces"
LIST=( $(func) )
show

echo "Tuned read: not good for appends to array"
IFS=$'\n' read -r -d '' -a LIST < <(func)
IFS=$'\n' read -r -d '' -a LIST < <(func2)
show

echo "Custom helper: good -- we will append to last listed but tested to work if there no array yet"
append_to_array LIST func
show

echo "Trying more complicated case with a pipe"
LIST=()
append_to_array LIST < <(func | tr o X)
show

[rusq]

Hey @yarikoptic, thanks for your contribution!

[rusq]

LGTM

[rusq]

I have left some comments in the other PR that you have submitted, if you could have a look at it too: #271