Removing pandas-profiling from post-setup checks

Reasons:
- Pandas-profling has been remamed ydata-profiling
- We removed it in /pull/272
  from the Apple Silicon requirements,
  because of incompatibility with scipy

It is a non-essential package anyway, and is used
in only one challenge (04-01-03 Exploratory
Analysis), which includes instructions to pip
install it.

checks/pip_check.sh

@@ -1,4 +1,4 @@
-REQUIRED=('pytest' 'pylint' 'ipdb' 'PyYAML' 'nbresult' 'autopep8' 'flake8' 'yapf' 'lxml' 'requests' 'beautifulsoup4' 'jupyterlab' 'pandas' 'matplotlib' 'seaborn' 'plotly' 'scikit-learn' 'tensorflow' 'nbconvert' 'xgboost' 'statsmodels' 'pandas-profiling' 'jupyter-resource-usage')
+REQUIRED=('pytest' 'pylint' 'ipdb' 'PyYAML' 'nbresult' 'autopep8' 'flake8' 'yapf' 'lxml' 'requests' 'beautifulsoup4' 'jupyterlab' 'pandas' 'matplotlib' 'seaborn' 'plotly' 'scikit-learn' 'tensorflow' 'nbconvert' 'xgboost' 'statsmodels' 'jupyter-resource-usage')
 PACKAGES=$(pip freeze)
 PACKS=()
 while read -r line; do
@@ -14,7 +14,7 @@ elif [ "${arch_name}" = "arm64" ]; then
   arch_name='m1'
 fi
 if [ $arch_name = 'm1' ]; then
-  REQUIRED=('pytest' 'pylint' 'ipdb' 'PyYAML' 'nbresult' 'autopep8' 'flake8' 'yapf' 'lxml' 'requests' 'beautifulsoup4' 'jupyterlab' 'pandas' 'matplotlib' 'seaborn' 'plotly' 'scikit-learn' 'tensorflow-macos' 'nbconvert' 'xgboost' 'statsmodels' 'pandas-profiling' 'jupyter-resource-usage')
+  REQUIRED=('pytest' 'pylint' 'ipdb' 'PyYAML' 'nbresult' 'autopep8' 'flake8' 'yapf' 'lxml' 'requests' 'beautifulsoup4' 'jupyterlab' 'pandas' 'matplotlib' 'seaborn' 'plotly' 'scikit-learn' 'tensorflow-macos' 'nbconvert' 'xgboost' 'statsmodels' 'jupyter-resource-usage')
 fi
 for r in ${REQUIRED[@]}; do
   echo $r