Fix info parameter not appearing in autograding failure messages

[Copilot]

When expect_equal (and similar) failures include an info parameter, the "Failure message" column in grading results was blank. The TAP reporter emits multi-line messages as a YAML block scalar (|-), but the parser was calling trimws() on every collected line, destroying indentation — leaving raw = "|-" which renders as blank in a GitHub markdown table cell.

Changes

• Preserve line indentation when collecting msg_lines (removed trimws() from collection loop)
• Updated grep/sub patterns to match 2-space-indented YAML keys (^[[:space:]]*message:)
• Handle YAML block scalars: when message value is |- or |, read content lines until hitting a sibling key (^ [^[:space:]]), concatenating non-empty content with spaces
• Unified termination condition for both block-scalar and inline branches: stop at sibling YAML keys (^ [^[:space:]])

Before / After

# Test with info parameter:
expect_equal(answer_in_rmd, TRUE,
  info = "Make sure to include the number of rows in the .rmd file"
)

	Expectation	Failure message
✗	Ex 1: Rmd Exercise 1 reports correct number of rows	(blank) → `answer_in_rmd` is not TRUE Make sure to include the number of rows in the .rmd file

Original prompt

---

This section details on the original issue you should resolve

<issue_title>adding "info" to the unit test feedback</issue_title>
<issue_description>most of the tests have additional feedback under info. it doesn't show up in the result.

test_that("Ex 1: Rmd Exercise 1 reports correct number of rows", {
skip_if(length(.rmd_content) == 0)
data("datasaurus_dozen", package = "datasauRus", envir = environment())
solution_nrow <- as.character(nrow(get("datasaurus_dozen", envir = environment())))

potential_answers <- c(solution_nrow, "nrow\(datasaurus_dozen\)", "names\(datasaurus_dozen\)")

pattern <- paste0("(", paste(potential_answers, collapse = "|"), ")")

answer_in_rmd <- stringr::str_detect(.rmd_content, pattern) |> any()

expect_equal(answer_in_rmd, TRUE,
info = "Make sure to include the number of rows in the .rmd file"
)
})

the output looks like this, but the | Failure message is blank instead of using the info

LAB11 autograding results

7 of 47 expectations passing — 15/100

  | Expectation | Failure message -- | -- | -- ☑ | ok 1 Rmd file contains headers |   ☑ | ok 2 Rmd file contains custom header elements |   ☑ | ok 3 Rmd contains a minimum number of R code chunks |   ☑ | ok 4 R code chunks contain actual code (not all empty) |   ✗ | not ok 1 Ex 1: Titanic data object exists |   ❔ | ok 2 # SKIP Reason: No Titanic data object found |   ❔ | ok 3 # SKIP Reason: No Titanic data object found |   ❔ | ok 4 # SKIP Reason: No Titanic data object found |   ❔ | ok 5 # SKIP Reason: No Titanic data object found |   ❔ | ok 6 # SKIP Reason: No Titanic data object found |   ✗ | not ok 7 Ex 1: titanic3 data object exists |   ✗ | not ok 8 Ex 1: m_apparent model object exists |   ❔ | ok 9 # SKIP Reason: m_apparent not found |   ❔ | ok 10 # SKIP Reason: m_apparent not found |   ❔ | ok 11 # SKIP Reason: m_apparent not found |   ❔ | ok 12 # SKIP Reason: p_apparent not found |   ❔ | ok 13 # SKIP Reason: m_apparent not found |   ✗ | not ok 14 Ex 10: cv_confusion data frame exists |   ❔ | ok 15 # SKIP Reason: cv_confusion not found |   ❔ | ok 16 # SKIP Reason: cv_confusion not found |   ❔ | ok 17 # SKIP Reason: cv_confusion not found |   ❔ | ok 18 # SKIP Reason: cv_confusion not found |   ❔ | ok 19 # SKIP Reason: cv_confusion not found |   ❔ | ok 20 # SKIP Reason: cv_confusion not found |   ✗ | not ok 21 Ex 10: Rmd contains confusion matrix code (TP, FP, TN, FN) |   ✗ | not ok 22 Ex 10: Rmd discusses sensitivity vs specificity trade-off |   ✗ | not ok 23 Ex 2: logistic regression mo...

• Fixes adding "info" to the unit test feedback #17

---

⌨️ Start Copilot coding agent tasks without leaving your editor — available in VS Code, Visual Studio, JetBrains IDEs and Eclipse.