[feat] FU-level fusion

[HobbitQia]

This pass implements the FU-level fusion after DFG-level fusion (PR 194), which aims to find the minimum FU cost that can cover all patterns extracted in previous passes (wrapped in fused_op).

The algorithm can be depicted as below:

1. Pattern Extraction: Extracts fused operation patterns from the module and linearizes them via topological sort.
2. Standalone Operation Extraction: Collects standalone operations not inside fused patterns for hardware coverage.
3. Template Creation: Greedily merges patterns into shared hardware templates using cost-based accommodation with DFS mapping search.
4. Connection Generation: Generates optimized FU connections based on pattern dependencies with bypass support.
5. Execution Plan Generation: Creates parallel execution stages by grouping operations at the same topological level.
6. JSON Output: Writes hardware configuration including templates, connections, and execution plans to JSON file.

TODO:

• Replace FunctionalUnit with FunctionUnit in Architecture.h.
• Update pattern_name of fused_op so that this attribute better illustrates the structure of the pattern.

Output JSON File

Field	Description
template_id	Unique identifier for this template
instance_count	Number of instances each pattern
supported_single_ops	Individual operations this template can execute standalone
supported_composite_ops	Fused patterns this template can execute
functional_units	Array of FU definitions with their supported operations
fu_connections	Data routing paths between FUs
pattern_execution_plans	Detailed execution schedules for each pattern

---

Execution Plan Fields

Field	Description
pattern_id	Pattern being executed
pattern_name	pattern name
fu_mapping	Maps operation index to slot ID: [op0→fu0, op1→fu1, ...], e.g. [1, 2] means we will use fu1 and fu2 to execute op0 and op1 of this pattern.
execution_stages	Ordered stages of execution
parallel_fus	Slots executing in this stage (can be multiple for parallel ops)
parallel_ops	Operations executing in this stage

---

Example

        "pattern_execution_plans": [
          {
            "pattern_id": 1,
            "pattern_name": "fused_op:icmp->grant_predicate->grant_predicate",
            "fu_mapping": [0, 1, 2],
            "execution_stages": [
              {
                "stage": 0,
                "parallel_fus": [0],
                "parallel_ops": ["neura.icmp"]
              },
              {
                "stage": 1,
                "parallel_fus": [1, 2],
                "parallel_ops": ["neura.grant_predicate", "neura.grant_predicate"]
              }
            ]
          },

The diagram below shows Hardware Template 1, where FU 0 serves as the data source for both FU 1 and FU 2.

       Template 1 Pipeline
      =====================
      
           [ FU 0 ]  <-- (neura.icmp)
              |
      ________|________
     |                 |
  [ FU 1 ]          [ FU 2 ]  <-- (neura.grant_predicate)
     |                 |
 (Grant A)         (Grant B)