feat(controllers): add exponential backoff rate limiter to all contro…

[geoffrey1330]

…llers

Add exponential backoff rate limiter to all controllers

Fixes #138.

All 18 controllers were requeueing on errors with fixed intervals, causing
thundering-herd retry storms against the backend when it was degraded. When
every controller fails at the same time, they all retry at the same wall-clock
time — keeping sbcli saturated and delaying recovery.

This wires an exponential backoff rate limiter (base 5 s, max 5 min) into
every controller's SetupWithManager via a shared defaultControllerOptions()
helper. When a controller returns ctrl.Result{}, err, successive retries for
the same object back off to 10 s, 20 s, 40 s … up to 5 min, then reset once
the object reconciles successfully.

RequeueAfter returns used for intentional polling (waiting for node online,
drain slot checks, sync intervals) are unaffected.

Fix: prevent duplicate backend calls when status patches fail after mutations

A subtle risk introduced by exponential backoff: if a backend POST succeeds
(e.g. trigger migrate, promote, suspend) but the subsequent status patch fails,
returning ctrl.Result{}, err causes the controller to back off — then retry
with Triggered=false still in status, firing the same backend call again.

Status patches that follow a successful backend mutation now return
ctrl.Result{Requeue: true}, nil instead of ctrl.Result{}, err. This
requeues immediately without triggering backoff, retrying only the status patch
until it lands. Once Triggered=true is persisted the controller moves forward
normally — no duplicate calls.

Applied across StorageNodeOps, SimplyblocksStorageCluster, and
VolumeMigration controllers.

Test Result

user@users-MacBook-Pro-2 ~ % kubectl -n simplyblock get storagecluster                           
NAME                  STATUS   UUID                                   CONFIGURED   AGE
simplyblock-cluster   active   38c2c881-afdc-4b95-813b-81ff2b15d009   true         6m27s
user@users-MacBook-Pro-2 ~ % kubectl -n simplyblock get storagenodesets
NAME               TOTAL   ONLINE   AGE
simplyblock-node   3       3        6m33s
user@users-MacBook-Pro-2 ~ % kubectl -n simplyblock get storagenode    
NAME                      WORKER                          SOCKET   NODEIDX   UUID                                   STATUS   HEALTH   AGE
simplyblock-node-9cxvxo   vm03.simplyblock3.localdomain   0        0         3fd8eb7c-ba8b-41b6-9d0e-9e011cd1b16a   online   true     5m11s
simplyblock-node-mk7iwn   vm04.simplyblock3.localdomain   0        0         74c99f22-dd3c-4df5-b1cd-4e0a540a915c   online   true     4m30s
simplyblock-node-tc85nl   vm02.simplyblock3.localdomain   0        0         df2486fc-f8be-41a3-a97f-38a3c430f2ae   online   true     6m35s