feat: Add multi-statement transaction support

Implement PEP 249-compliant transaction control with extensions for
manual commit/rollback operations. This enables atomic multi-table
operations with REPEATABLE_READ isolation semantics.

Core API additions:
- connection.autocommit property for enabling/disabling auto-commit mode
- connection.commit() to commit active transactions
- connection.rollback() to rollback active transactions
- connection.get_transaction_isolation() returns current isolation level
- connection.set_transaction_isolation() validates isolation level
- TransactionError exception for transaction-specific failures

Implementation details:
- Added autocommit state caching in Session with optional server query
- Added TRANSACTION_ISOLATION_LEVEL_REPEATABLE_READ constant
- All transaction operations include proper error handling and telemetry
- Supports fetch_autocommit_from_server connection parameter

Testing:
- Unit tests covering all transaction methods and error scenarios
- e2e integration tests validating transaction behavior including
  multi-table atomicity, sequential transactions, and isolation semantics

Documentation:
- Comprehensive TRANSACTIONS.md guide with examples and best practices
- Updated README.md with basic usage and reference to detailed docs

Requires MST-enabled Databricks SQL warehouse and Delta tables with
'delta.feature.catalogOwned-preview' table property.

Author: jayantsing-db

README.md

@@ -67,6 +67,37 @@ or to a Databricks Runtime interactive cluster (e.g. /sql/protocolv1/o/123456789
 > to authenticate the target Databricks user account and needs to open the browser for authentication. So it 
 > can only run on the user's machine.
 
+## Transaction Support
+
+The connector supports multi-statement transactions with manual commit/rollback control:
+
+```python
+import os
+from databricks import sql
+
+connection = sql.connect(
+    server_hostname=os.getenv("DATABRICKS_HOST"),
+    http_path=os.getenv("DATABRICKS_HTTP_PATH")
+)
+
+# Disable autocommit to use explicit transactions
+connection.autocommit = False
+
+cursor = connection.cursor()
+try:
+    cursor.execute("INSERT INTO table1 VALUES (1, 'a')")
+    cursor.execute("INSERT INTO table2 VALUES (2, 'b')")
+    connection.commit()  # Commit both inserts atomically
+except Exception as e:
+    connection.rollback()  # Rollback on error
+    raise
+
+cursor.close()
+connection.close()
+```
+
+For detailed information about transaction behavior, isolation levels, error handling, and best practices, see [TRANSACTIONS.md](TRANSACTIONS.md).
+
 ## SQLAlchemy
 Starting from `databricks-sql-connector` version 4.0.0 SQLAlchemy support has been extracted to a new library `databricks-sqlalchemy`.