[WIP][annotation, proteomics] Refine annotation api interface, proteomic api join interface, add ability to join somascan data

[akotlar]

• Refines annotation dataset: melt_by_field => explode_field, melt_by_samples => melt_samples, and improve docstring explanations for both. These names more accurately describe what these arguments do
• Refine proteomic join interface, and TandemMassTagDataset: TandemMassTagDataset class is now a msgspec.Struct, and we do not force drop fields. We also now do not force rename fields, besides Index => gene_name; this may be restored in the future as we support additional variations of Frappe datasets
• Enable ability to join on either gene_name or ProteinID columns.
• Improve the example proteomic notebook to better demonstrate melting_samples and explode_field

WIP because we still need to add the function to join on somascan data, and expand tests.

[akotlar]

closing in favor of #515

With Github's implementation of squash merge, rebasing is too complex