Add versioning scheme and fix GitHub issues

[renatolfc]

In this PR, we introduce a versioning scheme and start syncing our docker image
tags to that version.

Doing so will prevent breaking users' clusters when we introduce breaking
changes to the back-end images.

Additionally, we now introduce the update command, which both updates the
vibe-core library in Python, and also syncs images in the local cluster.

We also fixed a DNS issue that could happen in some environments, which
prevented workflows from running, because the cluster would end up without
a proper resolver.

This PR fixes:

• Receiving Error: 'spaceeye.split' does not exist in the workflow graph when running visualize dataset notebook of "Crop Segmentation" #35;
• Error in running notebook crop_cycles.ipynb #40;
• Error in running notebook land_degradation.ipynb #41, and
• Received Input Validation Errors while running whatif.ipynb notebook for carbon sequestration. #42.

Further changes are:

Script

• [🎉 NEW] The update command supersedes the update-images command, and
  manages the Python library automatically
• [📈 IMPROVEMENT] We now use automatic retries by default when downloading
  binaries for cluster setup

Client

• [📈 IMPROVEMENT] We now show the total run duration in the monitor header

Workflows

• [🎉 NEW] We added a new workflow that downloads MODIS 8-day surface
  reflectance rasters that intersect with the input geometry and time range
• [🎉 NEW] We added a new workflow to download gNATSGO raster data that
  intersect with the input geometry and time range
• [📈 IMPROVEMENT] We added a step parameter to the chunk_onnx workflow to
  tune the step size of chunking
• [⚒️ FIX] We updated the linear_trend op to define the write_rel_limits
  attribute in RasterChunk

Notebooks

• [⚒️ FIX] We updated the crop_cycle notebook to use the step parameter to
  prevent errors when computing cycles with small regions
• [⚒️ FIX] We fixed the land_degradation notebook with our fix to the
  linear_trend op
• [⚒️ FIX] We updated the scenario.json file to match changes to the
  COMET-Farm API in the carbon whatif notebook
• [⚒️ FIX] We updated the wf_viz_crop_seg.yaml file with the correct sink name for the SpaceEye task
• [📈 IMPROVEMENT] The carbon whatif notebook now uses the monitor, instead of
  block_until_complete
• [📈 IMPROVEMENT] We ported the DeepMC notebook from Tensorflow to PyTorch

Backend

• [📈 IMPROVEMENT] We now add traceparent
  headers to our cloudevents, enabling better inspection of messaging in our
  clusters
• [📈 IMPROVEMENT] We now only deploy services in the cluster after images
  have been downloaded, making cluster creation 20% faster in our local
  tests
• [📈 IMPROVEMENT] We added an in-memory reader for ChipDataset
• [📈 IMPROVEMENT] We centralized hash computation, and improved our
  messaging system to pass cache information in messages, paving the way for
  improved caching mechanisms

Documentation

• [🎉 NEW] We added a new entry to the troubleshooting guide describing what
  to do when your containers run out of space

Co-authored-by: Alex Crown acrown@microsoft.com
Co-authored-by: Bruno Silva brunosilva@microsoft.com
Co-authored-by: Eduardo Rodrigues edrodrigues@microsoft.com
Co-authored-by: Naga Bilwanth Gangarapu Naga@zensa.co
Co-authored-by: Nalin Singal nasingal@microsoft.com
Co-authored-by: Rafael Padilha rpadilha@microsoft.com
Co-authored-by: Renato Luiz de Freitas Cunha renato.cunha@microsoft.com