Enhance auto selection of FieldPaths with filters on field names

[cvauclair]

Is your feature request related to a problem? Please describe. Yes! Currently, Sugrounds supports auto-selection of FieldPaths. This means that if you use one of the querying methods (e.g.: sg.query_df) with a FieldPath that does not select a scalar value, then Subgrounds will automatically grab all scalar fields available from that partial selection (see code snippet below).

However, especially during the data exploration phase, this is too much data to be useful (users have to go through dozens of columns to find the data that interests them. The other option, to individually select all fields of interest, is tedious and verbose. What is missing is a middle ground between both cases.

Introducing, filtered auto-selections! (not to be confused with filters on selections, i.e.: with the where argument). The idea is simple, it would work the same as the regular autocompletion of FieldPaths, while allowing users to specify a filter for the fields they want to select.

Describe the solution you'd like

from subgrounds import Subgrounds

sg = Subgrounds()

subgraph = sg.load_subgraph("https://api.thegraph.com/subgraphs/name/messari/uniswap-v3-ethereum")

# Select individual fields
sg.query_df([
  subgraph.Query.financialsDailySnapshots.timestamp,
  subgraph.Query.financialsDailySnapshots.totalValueLockedUSD,
])

# Auto-select all fields
sg.query_df([
  subgraph.Query.financialsDailySnapshots
])

# NEW: Select all fields starting with `cumulative`
sg.query_df([
  subgraph.Query.financialsDailySnapshots._select("cumulative*")
])

Describe alternatives you've considered None.

Additional context For consistency, subgraph.Query.financialsDailySnapshots._select("*") should be equivalent to selecting all the fields (the second case).

Implementation checklist

• [ ] Update FieldPath._select to be able to handle widlcards