perf(runner): build the provider map once per file, not twice

[Chemaclass]

🤔 Background

Last per-file redundancy from the fork census (#801-#815): the data-provider map was awk-scanned twice per test file — once in the header's test-counting $(...) subshell (whose cache dies with it) and again by the runner.

💡 Changes

• Build the map in the counting loop's own shell before the per-file subshell; the header reads the total via a return slot instead of a $(...) capture, so the cache survives into the runner.
• 3 → 2 awk forks per single-file run (all ~258 nested acceptance runs are single-file); multi-file runs unchanged except the last file. find_total_tests' stdout contract kept.
• Fork-budget regression test tightened to the new budget.

[Chemaclass]

🤔 Background

Last per-file redundancy from the fork census (#801-#815): the data-provider map was awk-scanned twice per test file — once in the header's test-counting $(...) subshell (whose cache dies with it) and again by the runner.

💡 Changes

• Build the map in the counting loop's own shell before the per-file subshell; the header reads the total via a return slot instead of a $(...) capture, so the cache survives into the runner.
• 3 → 2 awk forks per single-file run (all ~258 nested acceptance runs are single-file); multi-file runs unchanged except the last file. find_total_tests' stdout contract kept.
• Fork-budget regression test tightened to the new budget.

[Chemaclass]

🤔 Background

Last per-file redundancy from the fork census (#801-#815): the data-provider map was awk-scanned twice per test file — once in the header's test-counting $(...) subshell (whose cache dies with it) and again by the runner.

💡 Changes

• Build the map in the counting loop's own shell before the per-file subshell; the header reads the total via a return slot instead of a $(...) capture, so the cache survives into the runner.
• 3 → 2 awk forks per single-file run (all ~258 nested acceptance runs are single-file); multi-file runs unchanged except the last file. find_total_tests' stdout contract kept.
• Fork-budget regression test tightened to the new budget.

[Chemaclass]

🤔 Background

Last per-file redundancy from the fork census (#801-#815): the data-provider map was awk-scanned twice per test file — once in the header's test-counting $(...) subshell (whose cache dies with it) and again by the runner.

💡 Changes

• Build the map in the counting loop's own shell before the per-file subshell; the header reads the total via a return slot instead of a $(...) capture, so the cache survives into the runner.
• 3 → 2 awk forks per single-file run (all ~258 nested acceptance runs are single-file); multi-file runs unchanged except the last file. find_total_tests' stdout contract kept.
• Fork-budget regression test tightened to the new budget.

[Chemaclass]

🤔 Background

Last per-file redundancy from the fork census (#801-#815): the data-provider map was awk-scanned twice per test file — once in the header's test-counting $(...) subshell (whose cache dies with it) and again by the runner.

💡 Changes

• Build the map in the counting loop's own shell before the per-file subshell; the header reads the total via a return slot instead of a $(...) capture, so the cache survives into the runner.
• 3 → 2 awk forks per single-file run (all ~258 nested acceptance runs are single-file); multi-file runs unchanged except the last file. find_total_tests' stdout contract kept.
• Fork-budget regression test tightened to the new budget.

[Chemaclass]

🤔 Background

Last per-file redundancy from the fork census (#801-#815): the data-provider map was awk-scanned twice per test file — once in the header's test-counting $(...) subshell (whose cache dies with it) and again by the runner.

💡 Changes

• Build the map in the counting loop's own shell before the per-file subshell; the header reads the total via a return slot instead of a $(...) capture, so the cache survives into the runner.
• 3 → 2 awk forks per single-file run (all ~258 nested acceptance runs are single-file); multi-file runs unchanged except the last file. find_total_tests' stdout contract kept.
• Fork-budget regression test tightened to the new budget.

[Chemaclass]

🤔 Background

Last per-file redundancy from the fork census (#801-#815): the data-provider map was awk-scanned twice per test file — once in the header's test-counting $(...) subshell (whose cache dies with it) and again by the runner.

💡 Changes

• Build the map in the counting loop's own shell before the per-file subshell; the header reads the total via a return slot instead of a $(...) capture, so the cache survives into the runner.
• 3 → 2 awk forks per single-file run (all ~258 nested acceptance runs are single-file); multi-file runs unchanged except the last file. find_total_tests' stdout contract kept.
• Fork-budget regression test tightened to the new budget.