Get rid of succ & fail marks on rate summary

[joanlopez]

What?

It replaces the use of succ & fail marks (✓/✗) for rate metrics on summary, in favor of the text: {success} out of {total}.

Note: the check summary remains the same, as for checks there's no ambiguity on the use of ✓ and ✗.

Why?

Because as discussed in #2306 and #3656, the current form is sometimes confusing, because the metric may have either a "positive" meaning (like successful requests) or a "negative" meaning (like failed requests), for which the use of ✓ may result in confusions, especially for the latter.

Checklist

• I have performed a self-review of my code.
• I have added tests for my changes.
• I have run linter locally (make lint) and all checks pass.
• I have run tests locally (make tests) and all tests pass.
• I have commented on my code, particularly in hard-to-understand areas.

Related PR(s)/Issue(s)

Closes #2306

Second iteration of #3656, based on what was discussed there.

[oleiade]

This PR aligns with our goal and designs to refactor the end-of-test summary. As a matter of fact, we integrated its approach in our designs. From my perspective, this is good to go and merge, in principle 👍🏻

[joanlopez]

We (@oleiade and I, among others) have been discussing details like what's mentioned in #3765 (comment), and others covered in the design doc, like grouping metrics (http, network, execution, etc) and we agreed to merge this, as it already brings some value to those users that still nowadays are getting confused with the current format using ✓ and ✗ for rate metrics (see this example), and is compatible with changes that will come in the near future.

So, we ship some value in the next/current release, and iterate later, instead of waiting till the end result to ship the value.

Besides, I'm not adding the breaking-change label, as we don't consider this summary format to be part of the expectations associated with version compatibility, while others could (e.g. machine readable ones, like JSON).