Test case for changing backend hash during aborted state migration

[lsierant]

Failing test case for issue #26259

Also, I tried to come up with a solution, but there are multiple cases when dealing with state migrations and that defeated me unfortunately.

Simple fix by removing this early state update in 'backend_C_r_S_unchanged ' will not work as it brakes init_test.go:TestInit_backendReinitConfigToExtra.

func (m *Meta) backend_C_r_S_unchanged
[...]
	// it's possible for a backend to be unchanged, and the config itself to
	// have changed by moving a parameter from the config to `-backend-config`
	// In this case we only need to update the Hash.
	if c != nil && s.Backend.Hash != uint64(cHash) {
		s.Backend.Hash = uint64(cHash)
		if err := sMgr.WriteState(s); err != nil {
			diags = diags.Append(err)
			return nil, diags
		}
	}
[...]

What I have found so far is that m.backend_C_r_S_changed casually calls m.backend_C_r_S_unchanged which updates hash only:

func (m *Meta) backend_C_r_S_changed(c *configs.Backend, cHash int, sMgr *clistate.LocalState, output bool) (backend.Backend, tfdiags.Diagnostics) {
// [...]
        // Grab the existing backend
	oldB, oldBDiags := m.backend_C_r_S_unchanged(c, cHash, sMgr)
// [...]

[github-actions]

I'm going to lock this pull request because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active contributions.
If you have found a problem that seems related to this change, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.