added coresight helper(utils) and sink-source-test

[Rohan-in-Qualcomm]

• Coresight-Sink-Source-Test: Validate trace capture across all CoreSight source and sink combinations by enabling sources, routing to sinks, and verifying collected trace data.

Also a common library helper function coresight_helper.sh is added to prevent duplication of code blocks.

[smuppand]

@Rohan-in-Qualcomm This PR is too large for a clean review in its current form.

It is 1 commit / 19 files and bundles a shared helper plus 6 different CoreSight suites. It also mixes repeated YAML/orchestration issues with runner-specific correctness bugs, which makes it hard to review and hard to rework safely.

Please split this into smaller PRs. Suggested grouping:

1. Runner/utils/coresight_helper.sh + Coresight-Sink-Source-Test
2. Ftrace-Dump-ETF-Base
3. Reset-All-Sinks-Base + Reset-Random-Sinks-Base
4. Single-Sink-Reset-Base + Single-Sink-Reset-Connected-Base

Also please update the PR title/body to match the actual scope. This is a CoreSight debug-suite PR, not an SMMU PR.

After the split, the file-level issues will be much easier to review cleanly.

[smuppand]

Thanks for adding the CoreSight helper and the sink/source validation test.

Since there are multiple pending CoreSight/QDSS PRs in parallel (#306, #370, #371, #392, #393, #394), I think we should treat #389 as the base/helper PR and review the rest as a coherent helper-based series.

Request:

1. Please make #389 establish the common CoreSight helper API that other tests can reuse.
2. Common operations such as debugfs/tracing path discovery, source/sink discovery, enable/disable, trace cleanup, sink reset, state restore, and result logging should live in the shared helper instead of being duplicated in each test.
3. Each test should have cleanup traps for EXIT/INT/TERM so enabled sources/sinks are disabled and trace paths are restored even if the test fails or is interrupted.
4. Missing CoreSight/debugfs/trace nodes should result in SKIP, not FAIL, unless the test explicitly targets a mandatory platform feature.
5. Please ensure scripts follow the existing repo conventions: init_env, functestlib.sh logging, check_dependencies, .res generation, POSIX shell, ShellCheck clean, and LAVA-friendly exit behavior.
6. Please include LAVA/manual validation logs for at least one supported target showing the helper and sink/source test working end to end.

Once this helper API is agreed, the dependent PRs (#392/#393/#394) and the related QDSS/QPMDA/TGU/ETM PRs (#370/#371/#306) should rebase on top of it and reuse the helper instead of carrying separate discovery/setup/cleanup logic.