handling project types and category mapping when data isn't in berkeley's database

[andersy005]

our current project type and category mapping rely on data from Berkeley's database. however, some registries (e.g., art-trees) aren't part of the database. as a result, the project categories are mapped to unknown

In [14]: df[(df.category == 'unknown')&(df.type != 'unknown')][['project_id', 'protocol', 'category', 'type', 'type_source']]
Out[14]: 
     project_id          protocol category           type type_source
3       VCS5460          [vm0047]  unknown  reforestation  carbonplan
5       VCS5462          [vm0047]  unknown  reforestation  carbonplan
9       VCS5446          [vm0047]  unknown  reforestation  carbonplan
15      VCS5433          [vm0047]  unknown  reforestation  carbonplan
19      VCS5417          [vm0047]  unknown  reforestation  carbonplan
21      VCS5412          [vm0042]  unknown    agriculture  carbonplan
26      VCS5404          [vm0047]  unknown  reforestation  carbonplan
28      VCS5403          [vm0047]  unknown  reforestation  carbonplan
29      VCS5401          [vm0047]  unknown  reforestation  carbonplan
31      VCS5399          [vm0047]  unknown  reforestation  carbonplan
32      VCS5398          [vm0047]  unknown  reforestation  carbonplan
37      VCS5389          [vm0047]  unknown  reforestation  carbonplan
39      VCS5392          [vm0042]  unknown    agriculture  carbonplan
41      VCS5387          [vm0047]  unknown  reforestation  carbonplan
42      VCS5386          [vm0047]  unknown  reforestation  carbonplan
46      VCS5382          [vm0047]  unknown  reforestation  carbonplan
48      VCS5376          [vm0042]  unknown    agriculture  carbonplan
76      VCS5333          [vm0047]  unknown  reforestation  carbonplan
78      VCS5324          [vm0047]  unknown  reforestation  carbonplan
84      VCS5316          [vm0047]  unknown  reforestation  carbonplan
92      VCS5304          [vm0042]  unknown    agriculture  carbonplan
95      VCS5299          [vm0042]  unknown    agriculture  carbonplan
107     VCS5282  [vm0047, vm0048]  unknown  reforestation  carbonplan
138     VCS5227          [vm0042]  unknown    agriculture  carbonplan
156     VCS5197          [vm0047]  unknown  reforestation  carbonplan
160     VCS5191          [vm0047]  unknown  reforestation  carbonplan
188     VCS5148          [vm0047]  unknown  reforestation  carbonplan
269     VCS5043          [vm0042]  unknown    agriculture  carbonplan
282     VCS5027          [vm0047]  unknown  reforestation  carbonplan
1628    VCS3354          [vm0042]  unknown    agriculture  carbonplan

the majority of the types above are from our infer_project_type function

@pf.register_dataframe_method
def infer_project_type(df: pd.DataFrame) -> pd.DataFrame:
    """
    Add project types to the DataFrame based on project characteristics
    Parameters
    ----------
    df : pd.DataFrame
        Input DataFrame containing project data.
    Returns
    -------
    pd.DataFrame
        DataFrame with a new 'type' column, indicating the project's type. Defaults to None
    """
    df.loc[:, 'type'] = 'unknown'
    df.loc[:, 'type_source'] = 'carbonplan'
    df.loc[df.apply(lambda x: 'art-trees' in x['protocol'], axis=1), 'type'] = 'redd+'
    df.loc[df.apply(lambda x: 'acr-non-fed' in x['protocol'], axis=1), 'type'] = (
        'improved forest management'
    )
    df.loc[df.apply(lambda x: 'vm0047' in x['protocol'], axis=1), 'type'] = 'reforestation'
    df.loc[df.apply(lambda x: 'vm0045' in x['protocol'], axis=1), 'type'] = (
        'improved forest management'
    )
    df.loc[df.apply(lambda x: 'vm0042' in x['protocol'], axis=1), 'type'] = 'agriculture'
    return df

how should we handle these cases?