feat: improve VCS waveform, coverage, and modern Verdi support

[SFangYy]

Description

This PR focuses on three main areas:

1. VCS waveform and coverage support
2. example output directory layout and per-example test entrypoints
3. xspcomm path resolution and generated-project linking behavior

The goal is to make the VCS flow more usable in real runs, make the example/test flows more consistent, and remove hardcoded assumptions about where xspcomm is installed.

Main changes

1. VCS waveform and coverage support

• add runtime control for VCS coverage save/reset flow
• make VCS finish handling safer for host-side callers
• stabilize waveform flushing and waveform filename rotation behavior
• add --verdi-mode to select between legacy Verdi integration and modern -debug_access+all
• only emit the Python-side libvcsnew.so visibility workaround for VCS when --verdi-mode modern is selected
• pass the Verdi mode through Python template rendering so generated Python DUTs do not fail on missing template variables

2. Example directory structure and test entrypoints

• unify generated example outputs under output/<ExampleName>
• add per-example make targets and align example scripts with the same output-root convention
• simplify local runs and CI flows by making example paths and entrypoints consistent

3. xspcomm path and linking fixes

• remove the hardcoded /usr/local/lib assumption from generated mem_direct builds
• resolve xspcomm paths through picker and use them in generated link/runtime settings
• improve portability for source-tree builds, custom install prefixes, and CI

Motivation / context

Before this PR:

• VCS waveform and coverage behavior was functional but fragile across different host/runtime setups
• example outputs and entrypoints were not fully consistent
• generated mem_direct projects assumed xspcomm lived in /usr/local/lib, which caused CI failures and made non-default installs unreliable
• the Python VCS flow could also fail when Verdi-mode-aware template logic was generated without the corresponding render variable

After this PR:

• VCS export flows are more explicit and robust
• example outputs and test targets are more uniform
• generated projects use detected xspcomm paths instead of a hardcoded system location

Fixes # (issue)

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

How Has This Been Tested?

Key validation performed for this change set:

• verified generated VCS/Python template rendering after adding __VERDI_MODE__ to Python codegen data
• reviewed the generated-project link path flow for mem_direct and confirmed it now uses picker-resolved xspcomm library paths instead of /usr/local/lib
• checked the example/test target restructuring against the current repository Makefiles and scripts

Suggested reproduction commands:

make test_vcs_RandomGenerator EXAMPLE_LANG=python
make test_RandomGenerator EXAMPLE_LANG=python
make test_InternalSignals EXAMPLE_LANG=python
make test_CacheSignalCFG EXAMPLE_LANG=python

Checklist:

• My code follows the style guidelines of this project
• I have added the appropriate labels
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in downstream modules

[Copilot]

Copilot wasn't able to review this pull request because it exceeds the maximum number of lines (20,000). Try reducing the number of changed lines and requesting a review from Copilot again.