viewFeedback page does not account for hooks

[damianhxy]

Steps To Reproduce
Add a parseAutoresult hook to the lab's config file that returns a score hash that differs from what would be obtained from a simple JSON.parse.

For example,

  def parseAutoresult(autoresult, _isOfficial)
    {"Problem 1": 1, "Problem 2": 2, "Problem 3": 3, "Problem 4": 4, "Problem 5": 5, "Problem 6": 6}
  end

Current behavior
viewFeedback page shows scores based on score JSON

Expected behavior
viewFeedback page shows the final scores (after any hooks)

[damianhxy]

This can be achieved by replacing parseScore with something like

def getScores(scores)
    score_hash = scores.map { |s|
      [s.problem.name, s.score]
    }.to_h
    if @jsonFeedback&.key?("_scores_order") == false
      @jsonFeedback["_scores_order"] = score_hash.keys
    end
    @total = 0
    score_hash.keys.each do |k|
      @total += score_hash[k].to_f
    rescue NoMethodError
      flash.now[:error] ||= ""
      flash.now[:error] += "The score for #{k} could not be parsed.<br>"
      flash.now[:html_safe] = true
    end
    score_hash["_total"] = @total
    score_hash
  end

and calling @scoreHash = getScores @submission.scores.includes(:problem)

However, there is currently a distinction between autograded scores and "manual" scores, where the sidebar displays differently for manually graded scores. If we want to maintain this distinction, we should check the grader_id and only set @scoreHash if it is not 0.

[damianhxy]

As an interesting side-effect of the current approach, JSON strings in manual feedback is parsed