fix: guard grep -v in 16 test.sh verifiers against set -e exit

Under set -e, `grep -v '^$'` returns exit code 1 when there are no
matches, killing the verifier script before the reward file is written.
This caused verifier_parse_error in sg_only mode where ALL_CHANGED can
be empty after clone-at-verify.

Also fix curl-security-review-001 Dockerfile.sg_only clone manifest:
wrong mirror name (sg-evals/curl/curl -> sg-evals/curl--09e25b9d).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: sjarmak

benchmarks/ccb_build/flipt-dep-refactor-001/tests/test.sh

@@ -111,7 +111,7 @@ ALL_CHANGED="$ALL_CHANGED
 $(git diff --name-only 2>/dev/null)
 $(git diff --cached --name-only 2>/dev/null)
 $(git ls-files --others --exclude-standard 2>/dev/null)"
-ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$')
+ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$' || true)
 
 OUTSIDE_SCOPE=0
 if [ -n "$ALL_CHANGED" ]; then

benchmarks/ccb_build/flipt-flagexists-refactor-001/tests/test.sh

@@ -120,7 +120,7 @@ ALL_CHANGED="$ALL_CHANGED
 $(git diff --name-only 2>/dev/null)
 $(git diff --cached --name-only 2>/dev/null)
 $(git ls-files --others --exclude-standard 2>/dev/null)"
-ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$')
+ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$' || true)
 
 OUTSIDE_SCOPE=0
 if [ -n "$ALL_CHANGED" ]; then

benchmarks/ccb_design/django-rate-limit-design-001/tests/test.sh

@@ -130,7 +130,7 @@ ALL_CHANGED="$ALL_CHANGED
 $(git diff --name-only 2>/dev/null)
 $(git diff --cached --name-only 2>/dev/null)
 $(git ls-files --others --exclude-standard 2>/dev/null)"
-ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$')
+ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$' || true)
 
 OUTSIDE_SCOPE=0
 if [ -n "$ALL_CHANGED" ]; then

benchmarks/ccb_design/flipt-protobuf-metadata-design-001/tests/test.sh

@@ -144,7 +144,7 @@ ALL_CHANGED="$ALL_CHANGED
 $(git diff --name-only 2>/dev/null)
 $(git diff --cached --name-only 2>/dev/null)
 $(git ls-files --others --exclude-standard 2>/dev/null)"
-ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$')
+ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$' || true)
 
 OUTSIDE_SCOPE=0
 if [ -n "$ALL_CHANGED" ]; then

benchmarks/ccb_fix/django-modelchoice-fk-fix-001/tests/test.sh

@@ -87,7 +87,7 @@ ALL_CHANGED="$ALL_CHANGED
 $(git diff --name-only 2>/dev/null)
 $(git diff --cached --name-only 2>/dev/null)
 $(git ls-files --others --exclude-standard 2>/dev/null)"
-ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$')
+ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$' || true)
 
 OUTSIDE_SCOPE=0
 if [ -n "$ALL_CHANGED" ]; then

benchmarks/ccb_fix/flipt-eval-latency-fix-001/tests/test.sh

@@ -118,7 +118,7 @@ ALL_CHANGED="$ALL_CHANGED
 $(git diff --name-only 2>/dev/null)
 $(git diff --cached --name-only 2>/dev/null)
 $(git ls-files --others --exclude-standard 2>/dev/null)"
-ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$')
+ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$' || true)
 
 OUTSIDE_SCOPE=0
 if [ -n "$ALL_CHANGED" ]; then

benchmarks/ccb_secure/django-audit-trail-implement-001/tests/test.sh

@@ -86,7 +86,7 @@ ALL_CHANGED="$ALL_CHANGED
 $(git diff --name-only 2>/dev/null)
 $(git diff --cached --name-only 2>/dev/null)
 $(git ls-files --others --exclude-standard 2>/dev/null)"
-ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$')
+ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$' || true)
 
 OUTSIDE_SCOPE=0
 if [ -n "$ALL_CHANGED" ]; then

benchmarks/ccb_secure/django-cross-team-boundary-001/tests/test.sh

@@ -74,7 +74,7 @@ ALL_CHANGED="$ALL_CHANGED
 $(git diff --name-only 2>/dev/null)
 $(git diff --cached --name-only 2>/dev/null)
 $(git ls-files --others --exclude-standard 2>/dev/null)"
-ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$')
+ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$' || true)
 
 OUTSIDE_SESSIONS=0
 if [ -n "$ALL_CHANGED" ]; then

benchmarks/ccb_secure/django-legacy-dep-vuln-001/tests/test.sh

@@ -122,7 +122,7 @@ ALL_CHANGED="$ALL_CHANGED
 $(git diff --name-only 2>/dev/null)
 $(git diff --cached --name-only 2>/dev/null)
 $(git ls-files --others --exclude-standard 2>/dev/null)"
-ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$')
+ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$' || true)
 
 OUTSIDE_SCOPE=0
 if [ -n "$ALL_CHANGED" ]; then

benchmarks/ccb_secure/django-policy-enforcement-001/tests/test.sh

@@ -31,7 +31,7 @@ fi
 # Get diff of new code
 DIFF=$(git diff HEAD~1..HEAD -- django/db/models/ 2>/dev/null || git diff -- django/db/models/ 2>/dev/null)
 ALL_CHANGED=$(git diff --name-only 2>/dev/null; git diff --cached --name-only 2>/dev/null; git ls-files --others --exclude-standard 2>/dev/null)
-ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$')
+ALL_CHANGED=$(echo "$ALL_CHANGED" | sort -u | grep -v '^$' || true)
 
 FUNC_SCORE=0
 POLICY_SCORE=0