Introduce NicknameSettingList for explicit efit tree cascade

[samc24]

Summary

• Adds NicknameSettingList: user-specifiable cascade of efit tree candidates. Tries each via open_tree, catches mdsExceptions.TreeFOPENR (tree missing) and FetchDataError (nested cascade exhausted), returns the first that opens.
• Resolver accepts lists (efit_nickname_setting=["efit21", "efit18", "analysis"]) and comma-separated CLI strings (--efit-tree "efit21,efit18,analysis"). Tokens that match registered keys (e.g. "disruption") dispatch to their NicknameSetting class inside the cascade.
• Refactors DisruptionNicknameSetting._cmod_nickname to delegate to the new primitive -- replaces the disruption_time is None proxy with real fallback driven by tree-open errors. D3D path unchanged (its rundb SQL query is real functionality, not just cascade).
• Phase 2 hook: MDSConnection.tree_nickname_cascades stores the originating NicknameSettingList so a follow-up PR can extend retry to data-path-missing errors inside an opened tree. Currently written for top-level NicknameSettingList only; nested cases (dict-wrapped, DisruptionNicknameSetting internals) are not yet surfaced.

Behavior change

For efit_nickname_setting="disruption" on CMOD, the legacy proxy disruption_time is None → return "analysis" is replaced with a real open_tree("efit18") attempt, falling back to "analysis" on TreeFOPENR.

Validated bit-identical output vs dev on two CMOD test shots:

• 1150805012 (disrupting flat-top): both branches resolve to efit18, 83 idx points, identical data.
• 1150805013 (no_disrup1_full, non-disrupting): old returns "analysis" via proxy; new tries efit18, gets TreeFOPENR, falls back to "analysis". Both produce 88 idx points, identical data.

Trade-off: one extra failed MDS open_tree per non-disrupting shot using "disruption", in exchange for a verbose log line documenting the fallback (Nickname cascade for shot N: selected 'analysis' after failed: efit18) and a path that handles unanticipated tree-availability patterns without proxy maintenance.

Out of scope (Phase 2 follow-up)

• Cascade retry on FetchDataError when a tree opens but a data path inside it is missing.
• Surfacing nested cascades (in NicknameSettingDict or inside DisruptionNicknameSetting._cmod_nickname) to MDSConnection.tree_nickname_cascades.
• Changing the default efit_nickname_setting value.

Test plan

• New unit tests in tests/test_nickname_setting.py (25 tests, pure-Python with a fake data_conn, no MDS required).
• Lint suite green: make black && ruff check . && isort --check . && pylint (CI run 26377445449).
• Manual CMOD smoke on mfews04 with shot 1150805012:
  • uv run disruption-py --efit-tree "doesnotexist,analysis" 1150805012 — cascade selects analysis after doesnotexist fails; verified via TRACE log.
  • uv run disruption-py --efit-tree "disruption,analysis" 1150805012 — first item resolves to DisruptionNicknameSetting, succeeds; 83 idx points (same as plain "disruption").
• Bit-identical regression check vs dev on shots 1150805012 (disrupting) and 1150805013 (non-disrupting) under --efit-tree "disruption".