[MLIR] Add `ngraph` prefix to MLIR flags (#3625)

* [MLIR] Add `ngraph` prefix to MLIR flags

Some flags collision with some MLIR flags.

* [MLIR] Add support for nGraph tensor type in parser

Initial commit that enables nGraph parsing. It's needed for testing.

* Rename ngraph print flag

* Rename ngraph dump mlir flags

* Clang format

* Revert "[MLIR] Add support for nGraph tensor type in parser"

This reverts commit ae371d6a5c8ea590322d5d3b9ba110159d4bf5fa.

src/contrib/mlir/compiler/compiler.cpp

@@ -85,7 +85,7 @@ using namespace ngraph::runtime::ngmlir;
 // *** Debug flags ***
 
 static llvm::cl::opt<bool> clPrintIRAfterAll(
-    "print-ngraph-ir-after-all",
+    "ngraph-print-ir-after-all",
     llvm::cl::init(false),
     llvm::cl::desc(
         "Print IR after transformation that are not implemented as passes in the MLIRCompiler. It "
@@ -99,37 +99,37 @@ static llvm::cl::opt<bool> clEnableNgInPlaceMemoryOpt(
     llvm::cl::desc("Enable ngraph dialect in-place memory optimization pass"));
 
 static llvm::cl::opt<bool>
-    clEnableAffineLoopFusion("affine-loop-fusion",
+    clEnableAffineLoopFusion("ngraph-affine-loop-fusion",
                              llvm::cl::init(false),
                              llvm::cl::desc("Enable loop fusion optimization in Affine dialect"));
 
 static llvm::cl::opt<bool>
-    clEnableAffineLoopTiling("affine-loop-tile",
+    clEnableAffineLoopTiling("ngraph-affine-loop-tile",
                              llvm::cl::init(false),
                              llvm::cl::desc("Enable loop tiling optimization in Affine dialect"));
 
 static llvm::cl::opt<unsigned>
-    clLoopTilingCacheLevel("affine-loop-tile-cache-level",
+    clLoopTilingCacheLevel("ngraph-affine-loop-tile-cache-level",
                            llvm::cl::init(2),
                            llvm::cl::desc("Cache level to which to apply affine loop tiling."));
 
 static llvm::cl::opt<unsigned> clLoopTilingCacheSize(
-    "affine-loop-tile-cache-size",
+    "ngraph-affine-loop-tile-cache-size",
     llvm::cl::init(0),
     llvm::cl::desc(
         "Cache size to use in affine loop tiling. If not zero, it overrides the cache-size "
         "inferred from the host CPU using for the cache level specified by "
-        "-loop-tile-cache-level."));
+        "-ngraph-loop-tile-cache-level."));
 
 // *** Debug flags ***
 
 static llvm::cl::opt<bool>
-    clDumpObjectFile("dump-mlir-object-file",
+    clDumpObjectFile("ngraph-dump-mlir-object-file",
                      llvm::cl::desc("Dump MLIR JITted-compiled object to file specified with "
                                     "-object-filename (<input file>.o by default)."));
 
 static llvm::cl::opt<std::string>
-    clObjectFilename("mlir-object-filename",
+    clObjectFilename("ngraph-mlir-object-filename",
                      llvm::cl::desc("Dump MLIR JITted-compiled object to file jitted_mlir.o"));
 
 #define COMPILE_OP_DECL(op_name)                                                                   \