signpost: cancel-upgrade behavior fix

[etungsten]

Issue number: Fixes #949

Description of changes:

Author: Erikson Tung <etung@amazon.com>
Date:   Fri Jun 19 11:55:10 2020 -0700

    signpost: `cancel-upgrade` behavior update
    
    Changes `cancel-upgrade` so that it only reverts actions done by
    `upgrade-to-inactive` as the usage information implies.
    `cancel-upgrade` now preserves the number of tries left on the
    inactive partition instead of just wiping all bits

Testing done:
Initial status:

bash-5.0# signpost status
OS disk: /dev/nvme0n1
Set A:   boot=/dev/nvme0n1p2 root=/dev/nvme0n1p3 hash=/dev/nvme0n1p4 priority=1 tries_left=0 successful=true
Set B:   boot=/dev/nvme0n1p6 root=/dev/nvme0n1p7 hash=/dev/nvme0n1p8 priority=0 tries_left=0 successful=false
Active:  Set A
Next:    Set A

Mark inactive partition as valid, see tries_left set to 1

bash-5.0# signpost mark-inactive-valid
bash-5.0# signpost status             
OS disk: /dev/nvme0n1
Set A:   boot=/dev/nvme0n1p2 root=/dev/nvme0n1p3 hash=/dev/nvme0n1p4 priority=1 tries_left=0 successful=true
Set B:   boot=/dev/nvme0n1p6 root=/dev/nvme0n1p7 hash=/dev/nvme0n1p8 priority=0 tries_left=1 successful=false
Active:  Set A
Next:    Set A

Make the inactive as next to boot, see priority bumped to 2

bash-5.0# signpost upgrade-to-inactive
bash-5.0# signpost status             
OS disk: /dev/nvme0n1
Set A:   boot=/dev/nvme0n1p2 root=/dev/nvme0n1p3 hash=/dev/nvme0n1p4 priority=1 tries_left=0 successful=true
Set B:   boot=/dev/nvme0n1p6 root=/dev/nvme0n1p7 hash=/dev/nvme0n1p8 priority=2 tries_left=1 successful=false
Active:  Set A
Next:    Set B

cancel-upgrade then properly rolls back upgrade-to-inactive by setting the priority back to 0 while keeping tries_left at 1

bash-5.0# signpost cancel-upgrade                                                                                                                                             
bash-5.0# signpost status        
OS disk: /dev/nvme0n1
Set A:   boot=/dev/nvme0n1p2 root=/dev/nvme0n1p3 hash=/dev/nvme0n1p4 priority=2 tries_left=0 successful=true
Set B:   boot=/dev/nvme0n1p6 root=/dev/nvme0n1p7 hash=/dev/nvme0n1p8 priority=0 tries_left=1 successful=false
Active:  Set A
Next:    Set A

Subsequent upgrade-to-inactive call is successful because the inactive partition is still marked valid (tries_left > 0)

bash-5.0# signpost upgrade-to-inactive
bash-5.0# signpost status             
OS disk: /dev/nvme0n1
Set A:   boot=/dev/nvme0n1p2 root=/dev/nvme0n1p3 hash=/dev/nvme0n1p4 priority=1 tries_left=0 successful=true
Set B:   boot=/dev/nvme0n1p6 root=/dev/nvme0n1p7 hash=/dev/nvme0n1p8 priority=2 tries_left=1 successful=false
Active:  Set A
Next:    Set B

cancel-upgrade then restores active partition as next to boot.
Another clear-inactive call then wipes the inactive partition priority bits.

bash-5.0# signpost cancel-upgrade     
bash-5.0# signpost status             
OS disk: /dev/nvme0n1
Set A:   boot=/dev/nvme0n1p2 root=/dev/nvme0n1p3 hash=/dev/nvme0n1p4 priority=2 tries_left=0 successful=true
Set B:   boot=/dev/nvme0n1p6 root=/dev/nvme0n1p7 hash=/dev/nvme0n1p8 priority=0 tries_left=1 successful=false
Active:  Set A
Next:    Set A
bash-5.0# signpost clear-inactive
bash-5.0# signpost status        
OS disk: /dev/nvme0n1
Set A:   boot=/dev/nvme0n1p2 root=/dev/nvme0n1p3 hash=/dev/nvme0n1p4 priority=2 tries_left=0 successful=true
Set B:   boot=/dev/nvme0n1p6 root=/dev/nvme0n1p7 hash=/dev/nvme0n1p8 priority=0 tries_left=0 successful=false
Active:  Set A
Next:    Set A
bash-5.0# 

Terms of contribution:

By submitting this pull request, I agree that this contribution is dual-licensed under the terms of both the Apache License, version 2.0, and the MIT license.

[etungsten]

Push above addresses @tjkirch 's comment.

[tjkirch]

Would like to make sure @iliana approves.