feat(rdb): allow actions to handle a full disk situation

[Mia-Cross]

Closes #2046

The provider didn't allow changes to node_type or volume_size_in_gb once the instance reached disk_full status. This is because both these attributes belong in the upgrades category and it tried to perform the updates before, which failed since the resource is in a transient state.

This PR fixes the issue by performing the upgrades first and adds a bit of documentation on how to deal with a disk_full situation with Terraform.

[codecov-commenter]

Codecov Report

Attention: 52 lines in your changes are missing coverage. Please review.

Comparison is base (1339bd5) 73.28% compared to head (d39302b) 73.23%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2209      +/-   ##
==========================================
- Coverage   73.28%   73.23%   -0.05%     
==========================================
  Files         212      212              
  Lines       33367    33388      +21     
==========================================
- Hits        24453    24452       -1     
- Misses       6817     6838      +21     
- Partials     2097     2098       +1     

Files	Coverage Δ
scaleway/resource_rdb_instance.go	70.67% <24.63%> (-1.31%)	⬇️

... and 2 files with indirect coverage changes

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