fix(initiator): volume expansion avoid target disconnect & sync dm size

[davidcheng0922]

Which issue(s) this PR fixes:

longhorn/longhorn#12359

What this PR does / why we need it:

• Option for not stopping target in StartNvmeTCPInitiator:

  During volume expansion, we avoid disconnect the target. Any disconnect / connect process may create orphan device.

• Add function SyncDmDeviceSize:

  Kernel takes time for getting the latest size, the dm table may stay previous size if we don't wait for it

Special notes for your reviewer:

Additional documentation or context

[codecov]

Codecov Report

❌ Patch coverage is 0% with 30 lines in your changes missing coverage. Please review.
✅ Project coverage is 0.00%. Comparing base (7df556b) to head (42d3288).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
pkg/initiator/initiator.go	0.00%	29 Missing ⚠️
app/cmd/nvmecli/nvmecli.go	0.00%	1 Missing ⚠️

Additional details and impacted files

@@          Coverage Diff          @@
##            main    #252   +/-   ##
=====================================
  Coverage   0.00%   0.00%           
=====================================
  Files         38      38           
  Lines       4874    4898   +24     
=====================================
- Misses      4874    4898   +24     

Flag	Coverage Δ
unittests	0.00% <0.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[derekbit]

default hard-coded value?

[davidcheng0922]

Investigate it

[davidcheng0922]

After investigating, it seems in linux kernel and dmsetup, it should always be 512, no matter the sector size of physical device.

/**
 * The type used for indexing onto a disc or disc partition.
 *
 * Linux always considers sectors to be 512 bytes long independently
 * of the devices real block size.
 *
 * blkcnt_t is the type of the inode's block count.
 */
typedef u64 sector_t;

ref: https://github.com/torvalds/linux/blob/master/include/linux/types.h#L130-L138

/* FIXME Should be elsewhere */
#define SECTOR_SHIFT 9L

ref: https://android.googlesource.com/platform/external/lvm2/+/refs/heads/main/tools/dmsetup.c#103

Do I still implement a helper function for this?

[davidcheng0922]

If I'm wrong, please correct me, thank you

[derekbit]

LGTM