fix(scheduler): fall back from PUT to GET on pause/resume 405

[kowser-orkes]

Pull Request type

• Bugfix
• Feature
• Refactoring (no functional changes, no api changes)
• Build related changes
• WHOSUSING.md
• Other (please describe):

Changes in this PR

• Schedule pause/resume verb differs by server family (OSS/embedded Conductor accepts PUT only,
  Orkes Conductor historically accepted GET only) — both SchedulerResourceApi (the generated
  client) and Schedules (the AI scheduling facade) sent a single hardcoded verb, so each only
  worked against one server family
• Both now try PUT first on every call, falling back to GET only on a 405 — no cached state, no
  shared verdict between pause and resume
• Any other error status (403, 404, 5xx) propagates immediately, no fallback attempted
• No public API changes — same method signatures, same return types, same exceptions on real
  failures

Alternatives considered

• Switch the hardcoded verb from GET to PUT outright — rejected, breaks any Orkes deployment
  that hasn't picked up server-side PUT support yet
• Cache the fallback verdict per client instance (matching the Java SDK client) — rejected in
  favor of a simpler, stateless per-call retry

[chrishagglund-ship-it]

lgtm