Changes for MVP-4632

[ubyndr]

• Added Library Documentation link to README.md.
• Added MissingAnalysisProcess exception to warn users to use any of the co_annotation_report method before using GraphGenerator.
• Refactored couple of test cases for the previous changes.
• Start nodes can already be selected with labels using node_selector. See explanations below;

def visualize_rdf_graph(
    self,
    predicate: Optional[str] = None,
    start_node: Optional[List[str]] = None,
    node_selector: Optional[Dict[str, str]] = None,
    file_path: Optional[str] = None,
    bottom_up: Optional[bool] = True,
):

node_selector: A dictionary specifying how to select a node when node URIs are not used.
    Defaults to None. The dictionary should have the following format:

    - To select by label:
        {'property': 'label', 'value': 'x label'}

    - To select by annotation:
        {'property': 'x_annotation', 'value': 'xxxx'}

    - The 'property' key represents the property that will be queried.
    - The 'value' key represents the desired property value to match.

[hkir-dev]

Some suggestions to improve walkthrough.ipynb

• At cell 22, can we update gg.visualize_rdf_graph step to use the node_selector with label? Because ids are auto generated and making it hard to do copy>paste>run at this step.
• At Cell 16 should we provide author_cell_type_list to AnndataEnrichmentAnalyzer constructor? It didn't worked without that parameter in my case.